# Orfloat — Forward Deployed Applied AI > Editorial: this file is the agent-readable surface of orfloat.com. > Human visitors should read https://www.orfloat.com instead. **Studio:** Orfloat **Parent entity:** Afraa & Mufassir LLC **Registration:** CR 1504141 · License L3717644 **Founded:** 2026 **Based in:** Muscat, Sultanate of Oman **Coverage:** GCC, with travel **Contact:** orfloat.studio@gmail.com **Stack:** Claude · MCP · Skills · Agents · Claude Code **Independence:** Orfloat is independent from Anthropic. We build on Anthropic's published primitives; we are not partnered with, endorsed by, or formally affiliated with Anthropic PBC. **Website:** https://www.orfloat.com **This file:** https://www.orfloat.com/llms-full.txt **Index of agent-readable pages:** https://www.orfloat.com/llms.txt --- # Orfloat — We follow the pioneer > Source: https://www.orfloat.com/ > Updated: 2026-05-31 # We follow the pioneer Anthropic is building the most capable, and most carefully aligned, AI of this era. We are not Anthropic. We follow in its footsteps, in Muscat, Oman, as an Applied AI engineering studio that takes the frontier it publishes and turns it into systems Omani enterprises can trust. We believe the gap that matters is not technical. It is the distance between what Claude can already do, and what we have the discipline and the courage to ask of it. So we hold our work to the standard the pioneer sets: its research, its alignment, its ethics. Not as decoration, but as the way the work is done. Capability without that is only a faster way to be wrong. It is a ledger of respect, and an invitation. The future is already here, built on Claude, and those with the eyes to see it will not need persuading. - [See the practice](https://www.orfloat.com/practice) - [Anthropic](https://www.anthropic.com) --- # Orfloat — The Practice > Source: https://www.orfloat.com/practice > Updated: 2026-05-31 # Orfloat **Forward deployed applied AI, for thoughtful businesses.** We embed engineers inside a handful of GCC businesses each year and turn what Claude can already do into systems your team will still trust at 2am on a Friday. - **Studio:** Afraa & Mufassir LLC - **Practice:** Applied AI · Forward Deployed - **Stack:** Claude · MCP · Skills · Agents - **Region:** Muscat & the wider GCC - **Engagement:** Typically 15 days on-site for Discovery [Start a Discovery Phase](https://www.orfloat.com/contact) · [Read the manifesto](https://www.orfloat.com/services) ## 01 — The Method: The 4D framework Four disciplines we hold every Orfloat engagement to — inspired by Anthropic's published Applied AI practice. They are the difference between an interesting demo and a system you can stake the business on. - **01 / Delegation.** Move the right work to the model. Not everything — only the tasks where Claude is faster, calmer, or more consistent than the human doing them today. - **02 / Description.** Spell out the work as if for a thoughtful new colleague. Context, constraints, evidence, the shape of a good answer. Tools and Skills are the vocabulary. - **03 / Discernment.** Read the output the way a senior reads a junior. Calibrate trust to the task. Build the eval before you build the agent. Notice drift early. - **04 / Diligence.** Stay in the loop. Treat safety, privacy, and reversibility as primary constraints — not afterthoughts. Operate the system, don't just deploy it. ## Six primitives, one coherent practice Built on Claude · Anthropic. Anthropic ships extraordinary primitives. Most teams use one of them. Orfloat composes all six into systems calibrated to your operation — and stays embedded long enough for the seams to disappear. - **Primitive 01 — Anthropic API.** Direct access to frontier reasoning. Claude Opus, Sonnet, and Haiku — calibrated to the latency, cost, and judgement each step of your workflow actually needs. - **Primitive 02 — Claude Code & SDKs.** Agentic coding embedded in your team's workflow. We deploy Claude Code with your repo, your conventions, and a curated set of guardrails specific to your codebase. - **Primitive 03 — Model Context Protocol.** A common language between Claude and your existing systems. Inventory, POS, ERP, CRM, calendars, sheets — exposed cleanly so the model reasons over real, current state. - **Primitive 04 — Skills.** Reusable, evaluated capabilities. The right Skill in scope, the wrong ones quietly out of the way — so the agent stays sharp on the work that matters today. - **Primitive 05 — Plugins & Tools.** Narrow, well-described tools the model can actually use. We design the contract, write the eval, and operate the integration once it's live. - **Primitive 06 — Claude-managed agents.** Persistent agentic loops with memory, retries, and human-in-the-loop checkpoints — for the work that doesn't finish in a single chat turn. ## 02 — Engagement model: Discovery first. Always. We will not propose a system before we have understood the work. Every Orfloat engagement begins with a fixed, time-boxed Discovery — because the alternative is shipping a beautiful answer to the wrong question. - **Phase 01 — Discovery.** Fifteen calendar days, on-site. We shadow the work end to end — front of house, back office, finance, supplier flows. You leave with an audit, an opportunity map, and a calibrated roadmap. *15 days · On-site.* - **Phase 02 — Service Agreement.** A scoped, milestone-based agreement built from Discovery — never a generic SOW. Effort, dependency, and impact are estimated from evidence, not guessed at. *Calibrated to your roadmap.* - **Phase 03 — Forward deployment.** Orfloat engineers sit inside the business while we ship. We build, evaluate, train your team to operate the system, and stay close through the first quarter of production. *Embedded · 90-day handover.* ## 03 — The Studio A boutique studio, embedded in the businesses we serve. - **Studio:** Orfloat - **Parent:** Afraa & Mufassir LLC - **Registration:** CR 1504141 · L3717644 - **Base:** Muscat, Sultanate of Oman - **Coverage:** GCC, with travel - **Cadence:** Selective intake ## 04 — Begin **Frameworks don't ship. *We do.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [See the work](https://www.orfloat.com/work) - Studio: Afraa & Mufassir LLC - Based in: Muscat, Oman - Stack: Built on Claude - Year: 2026 · Vol. 01 --- # Services > Source: https://www.orfloat.com/services > Updated: 2026-05-25 # Services **A narrow practice, *deeply done.*** Orfloat does one thing: forward-deployed Applied AI engineering for family-led and founder-run businesses across Oman and the wider GCC. Five offerings, all composable, all built on the same underlying stack: Claude and the Anthropic primitives that surround it. Every engagement runs against the three-dimensional model Anthropic publishes in their enterprise guide — People, Process, Technology — translated for the family-run scale we work at. ## 01 / Discovery **Operational discovery, on-site.** Fifteen days inside the business — shadowing the work, mapping the supply chain, sitting with the team. We leave with evidence, not assumptions. - End-to-end operational workflow audit - Bottleneck and friction map, with priority weights - Prioritised AI integration roadmap (effort × dependency × impact) - A draft Service Agreement, calibrated to what we found ## 02 / Customer-facing systems **Claude where your customer meets you.** Conversational agents that hold the brand voice. Intent capture that gets to specifics on the first turn. Returning-customer recall that feels considered, not surveilling. - WhatsApp / web agent for enquiries and bookings - Conversational intake of requirements and preferences - Returning-customer personalisation, owned by the business - AI-generated product, programme, and announcement copy - Automated review solicitation with quality gating ## 03 / Internal operations **Operational systems that don't go to sleep.** Inventory, supplier, scheduling, dashboards. The unglamorous work where Claude pays back daily — by removing the recurring questions, not the people. - Intelligent staff scheduling and shift optimisation - Inventory monitoring and supplier reorder triggers - AI-assisted operational dashboards and daily briefings - Workflow routing and demand forecasting - POS, booking, and accounting integrations (MCP-first) ## 04 / Brand & positioning **An AI-native posture, on your terms.** Most businesses in the region either ignore AI or wrap themselves in it badly. We help you build a confident, defensible narrative around how Claude actually shows up in your work. - Internal AI usage policy for staff - Customer-facing AI transparency principles - Brand voice guidance for AI touchpoints - Social and announcement automation, with editorial review ## 05 / Forward deployment **We stay until the seams disappear.** Orfloat engineers sit inside the business while the systems go live. We run the evals, watch the dashboards, and hand the keys over only when your team is operating it well. - Embedded engineering, weekly cadence - Evaluations designed before deployment - On-call coverage through the first 90 days - Team training and operating runbooks - Quarterly capability reviews as Claude evolves ## Begin **Five offerings. *One way to begin.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [How we work](https://www.orfloat.com/practice) --- # Work > Source: https://www.orfloat.com/work > Updated: 2026-05-25 # Work **Outcomes first. *Press second.*** A quiet, deliberate register. Orfloat publishes engagements only after the system has been in production long enough to talk about it honestly — and only with the client's explicit consent. ## The Register — 2026, Vol. 01 ### № 01 — Held in confidence. - **Status:** Discovery Phase ongoing (active) - **Sector:** Hospitality · Muscat - **Year:** 2026 — An active forward-deployed engagement with a hospitality group in Muscat. Client identity withheld at the client's request. A public case study will follow only after production milestones land and the client's written consent is in place. ### № 02 — Limited availability for new Discovery Phases. - **Status:** Currently accepting (open) - **Sector:** Any sector · GCC - **Year:** Open Orfloat takes on a small number of forward-deployment engagements at any one time. Each begins with a typically 15-day on-site Discovery Phase. If you are evaluating Claude seriously, we should talk. ### № 03 — Future engagements appear here as they ship. - **Status:** Held in confidence (reserved) - **Sector:** Reserved - **Year:** — Orfloat publishes client work only after the system has been in production long enough to talk about it honestly. Outcomes first. Press second. ### № 04 — Reserved for the next graduating engagement. - **Status:** Held in confidence (reserved) - **Sector:** Reserved - **Year:** — Forward deployment means the studio's bandwidth is genuinely finite. We would rather show one engagement we are proud of than several we are not. ## Begin **The next entry could be *yours.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [How we work](https://www.orfloat.com/practice) --- # Why now > Source: https://www.orfloat.com/why-now > Updated: 2026-05-29 # Why now **The frontier moves *weekly.*** Forecasting AI capability used to be the hard part. It isn't anymore. The hard part is reading what shipped this week, deciding which businesses it changes, and getting that change into production before the next thing ships. This page tracks our thinking against the moving frontier. ## 01 — The premise **Every business in Oman is operating against *a 2026 frontier* with a 2023 playbook.** Anthropic ships an upgrade every few weeks. Independent research changes monthly. The gap between "what Claude can already do" and "what most teams are using" has stopped narrowing and started widening. That is the central thesis of Orfloat — and the reason we publish notes here regularly. Not announcements. Working drafts of how we think about the frontier from the inside of a Muscat operating business. > The capability overhang is the most consequential under-discussed force in the Gulf economy today. Reading about AI on Twitter is not the same as having a roadmap. ## Latest field notes The three most recent notes — the full archive lives at [/notes](https://www.orfloat.com/notes). - **[Forty-two agents on our own codebase, and the call they couldn't make.](https://www.orfloat.com/notes/dogfooding-the-workflow)** — Field note · 29 May 2026 · 8 min. We pointed Claude Opus 4.8 and Claude Code's new dynamic workflows at our own website — 42 agents, ~14 minutes, one production deploy. The workflow surfaced a genuinely high-value fix and then nearly fooled itself. The judgment that caught it is the whole point. - **[Claude Opus 4.8, and the discipline it asks for.](https://www.orfloat.com/notes/opus-4-8-discipline)** — Field note · 28 May 2026 · 8 min. Anthropic shipped Opus 4.8 today — around four times less likely to let its own code flaws pass, paired with dynamic workflows that orchestrate hundreds of subagents from a single session. The capability stopped being the constraint a while ago. What is left is whether you can describe the work and discern the output — the disciplines, not the model. - **[Software after software, and the record so far.](https://www.orfloat.com/notes/software-after-software)** — Field note · 27 May 2026 · 9 min. Twelve theses on what software becomes when intelligence is abundant — and the empirical record from the last six months that says they are no longer speculative. [View the full archive](https://www.orfloat.com/notes) ## 02 — What we read **Our north star is *Anthropic's own thinking.*** We read [anthropic.com/news](https://www.anthropic.com/news) and [claude.com/blog](https://claude.com/blog) end-to-end as they ship. Around that, we follow frontier research from DeepMind, OpenAI, Meta FAIR, and the academic community at NeurIPS, ICML, and ACL. And we keep watch on regional signals — Vision 2040 progress, GCC enterprise adoption data, PwC and McKinsey's MENA work. When we publish a note, it's because we've read something that we think changes the decision a Muscat operator should be making this quarter. Not because we needed content for the site. ## Begin **Reading this is *not enough.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [Read all notes](https://www.orfloat.com/notes) --- # Why us > Source: https://www.orfloat.com/why-us > Updated: 2026-05-25 # Why us **Early. *Embedded.* Certified.** Most studios that talk about Claude have used it. Few have studied it the way Anthropic teaches it. Orfloat's CTO has completed every relevant course Anthropic Academy publishes — eight certifications spanning the Anthropic API, Claude Code (101 and the advanced "in Action" masterclass), Model Context Protocol (intro and advanced), Skills, subagents, and the business-side AI Fluency curriculum co-developed with PayPal. The studio operates on what those courses actually teach. ## 01 — Why this matters **There is a difference between *reading Anthropic* and operating to its standards.** Most teams stop at the marketing page. A few read the blog. Far fewer follow Anthropic's formal Academy curriculum — the courses they themselves use to train customer-facing engineers and partners on the right way to build with Claude. Every Orfloat engagement is delivered by a CTO who has completed that curriculum end-to-end, verifies the certificates publicly, and continues to ship side projects with each new primitive Anthropic releases. When you hire Orfloat, you're not hiring someone who finished a bootcamp two years ago. > We do not learn Claude from blog summaries. We do the work Anthropic ships, then we ship the work for clients. ## Anthropic Academy — Akram Ahmed, CTO Eight certificates of completion. ### Claude with the Anthropic API ![Anthropic Academy certificate of completion: Claude with the Anthropic API](https://www.orfloat.com/certs/claude-code-api-certificate@2x.webp) Direct API integration patterns — prompt engineering, system prompts, tool use, structured outputs, and streaming. The foundation of everything we ship. [Verify · h6bujppcv3tk](https://verify.skilljar.com/c/h6bujppcv3tk) ### Claude Code 101 ![Anthropic Academy certificate of completion: Claude Code 101](https://www.orfloat.com/certs/claude-code-101-certificate@2x.webp) Anthropic's official curriculum on Claude Code — how to embed it in a working repository with conventions, hooks, and guardrails specific to a codebase. Certificate of Completion ### Claude Code in Action ![Anthropic Academy certificate of completion: Claude Code in Action](https://www.orfloat.com/certs/claude-code-in-action-certificate@2x.webp) Anthropic's advanced Claude Code masterclass — real-world agentic coding workflows, autonomous task execution, multi-file refactors, and the production patterns we deploy day-to-day inside client repositories. [Verify · 6uqtggo7uw5y](https://verify.skilljar.com/c/6uqtggo7uw5y) ### Introduction to agent Skills ![Anthropic Academy certificate of completion: Introduction to agent Skills](https://www.orfloat.com/certs/skill-certificate@2x.webp) Building reusable, evaluated capabilities. The right skill in scope, the wrong ones quietly out of the way — so the agent stays sharp on the work that matters today. Certificate of Completion ### Introduction to subagents ![Anthropic Academy certificate of completion: Introduction to subagents](https://www.orfloat.com/certs/sub-agents-certificate@2x.webp) Decomposing long-running tasks into specialized sub-agents that coordinate via Claude — the architectural primitive behind systems that don't fall over at 2am. Certificate of Completion ### Introduction to Model Context Protocol ![Anthropic Academy certificate of completion: Introduction to Model Context Protocol](https://www.orfloat.com/certs/mcp-certificate@2x.webp) MCP fundamentals — resources, tools, prompts. The standard that lets Claude reason over your POS, ERP, and CRM at once without bespoke glue for each. [Verify · 2oz8uc26g6ab](https://verify.skilljar.com/c/2oz8uc26g6ab) ### Model Context Protocol: Advanced Topics ![Anthropic Academy certificate of completion: Model Context Protocol: Advanced Topics](https://www.orfloat.com/certs/mcp-advanced-certificate@2x.webp) Authentication, transport, server design, sandboxing. The depth that separates a working integration from a production one. [Verify · yv7rbkjdzrpo](https://verify.skilljar.com/c/yv7rbkjdzrpo) ### AI Fluency for Small Businesses ![Anthropic Academy certificate of completion: AI Fluency for Small Businesses](https://www.orfloat.com/certs/AI-SME-certificate@2x.webp) Co-developed by Anthropic and PayPal for small-business operators. The vocabulary, posture, and decision framework we use to brief non-technical founders inside our engagements. Certificate of Completion ## 02 — How we stay current **Continuous, not a credential.** The certificates above are not the work. They're the receipt for the work. The actual work is reading every Anthropic news post the day it ships, building against each new primitive in a personal sandbox before recommending it to clients, and following the developer community where the hard-won implementation lessons live. We also read Anthropic's long-form enterprise material end to end — the latest field note, [Anthropic's enterprise playbook, read from Muscat](https://www.orfloat.com/notes/anthropic-enterprise-playbook), works through their 35-page guide directly. Akram Ahmed (CTO) is a daily Claude Code user across Orfloat's own codebase — including this website. Mufassir Ahmed (CEO) leads the commercial side and the engagement with the wider Omani business community Orfloat exists to serve. ## Begin **A small number of businesses *will be early.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [About the studio](https://www.orfloat.com/about) --- # About > Source: https://www.orfloat.com/about > Updated: 2026-05-25 # About **A studio for the *capability overhang* era.** Orfloat is a software studio operating under Afraa & Mufassir LLC, registered in Muscat, Oman. We exist for a specific moment in the technology cycle — one in which the gap between what frontier models can do and what businesses actually use them for has never been wider, and has never mattered more. ## 01 — Why we exist **Most teams are using *a fraction* of what Claude can already do.** Anthropic ships extraordinary capability. APIs. SDKs. Model Context Protocol. Skills. Plugins. Claude-managed agents. A Constitution. A research culture that treats safety as primary. In Oman — and across most of the region — businesses are meeting that capability with a chat window and a few half-used prompts. The opportunity isn't to write more prompts. It's to embed. > Orfloat does forward-deployed Applied AI engineering. We sit inside the business, learn the work end to end, and ship systems that compound — built on Anthropic's primitives, with the same seriousness about evaluation, safety, and honesty we admire in their published practice. We are independent and not affiliated with Anthropic; we simply build on what they make. ## 02 — Principles **Seven commitments we hold ourselves to.** We borrow Anthropic's seriousness about safety, evaluation, and honesty — and we apply it at the scale of a small business in Muscat, not just a frontier lab in San Francisco. - **01 — Discovery before deployment.** We will not propose a system before we understand the work. Every engagement starts with a fixed Discovery. - **02 — Evals before agents.** If we cannot measure whether the system is working, we will not ship it. Period. - **03 — Human-in-the-loop by default.** Automation is a posture, not a goal. We design for reversible, supervised systems first. - **04 — Privacy and data dignity.** Client data is the client's. We design engagements with Oman's PDPL and Anthropic's Usage Policy in mind, and reflect that in every Engagement Letter. - **05 — Stay until it's true.** Forward deployment means we don't disappear at launch. The first 90 days of production are part of the work. - **06 — Tell the truth about AI.** Where Claude is the right answer, we say so. Where it is not, we say that too — even if it costs us the engagement. - **07 — Operate from Anthropic's published playbook.** We use the People–Process–Technology framework Anthropic publishes for enterprise deployments, plus their four-stage rollout and five LLMOps practices — translated for family-led businesses in Oman and the wider GCC. Attribution is built in. ## 03 — Founders **Two operators. *One thesis.*** Orfloat is run by two brothers. One ships software with Claude every day. The other runs the business that gives that work a home. Both believe Oman's small and mid-size enterprises deserve applied AI built with the same craft and discipline that frontier labs hold themselves to. ![Portrait of Akram Ahmed, Co-founder and CTO of Orfloat](https://www.orfloat.com/founders/akram@2x.webp) **Akram Ahmed — Co-founder · CTO.** Software Architect. Ships production systems with Claude Code daily, and leads engineering at Orfloat — system architecture, evaluations, and the day-to-day forward-deployed work inside client teams. ![Portrait of Mufassir Ahmed, Co-founder and CEO of Orfloat](https://www.orfloat.com/founders/mufassir@2x.webp) **Mufassir Ahmed — Co-founder · CEO · Director.** Director of Afraa & Mufassir LLC. Leads commercial, engagement, and client relationships across Orfloat's GCC operating footprint. ## 04 — The Studio **Muscat, *deliberately.*** - **Legal entity** — Afraa & Mufassir LLC - **Studio brand** — Orfloat - **Commercial Reg.** — CR 1504141 - **License** — L3717644 - **Address** — PO Box 1, P.C. 411, Sur, Oman - **Operating from** — Muscat, Sultanate of Oman - **Email** — orfloat.studio@gmail.com ## Begin **The principles are the promise. *Let us keep it.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [Why Orfloat](https://www.orfloat.com/why-us) --- # Contact > Source: https://www.orfloat.com/contact > Updated: 2026-05-25 # Contact **Start a *Discovery Phase.*** Every Orfloat engagement begins with a fixed, 15-day Discovery — on-site, embedded, evidence-based. Tell us a little about your business and we'll come back within two working days to schedule an introductory call. ## How to reach the studio - **Email** — - **Web form** — a contact form is available at the canonical [/contact](https://www.orfloat.com/contact) route for human visitors. Submitting the form does not create a contractual relationship; any engagement is subject to a signed Engagement Letter. Form submissions are processed in line with our [Privacy Notice](https://www.orfloat.com/privacy). ## Studio Afraa & Mufassir LLC Muscat, Sultanate of Oman ## Registration CR 1504141 · License L3717644 ## What to expect A short reply from one of the founders, an introductory call within the week, and — if both sides see a fit — an Engagement Letter for a typically 15-day on-site Discovery Phase. Replies come from **orfloat.studio@gmail.com** within two working days. Engagements are confidential at the client's request. See [what we actually do](https://www.orfloat.com/services) for the shape of the work. --- # Notes > Source: https://www.orfloat.com/notes > Updated: 2026-05-25 # Field notes from *the work.* Short essays from inside Orfloat engagements — on Claude, on forward deployment, and on what closing the capability overhang actually looks like at the scale of a Muscat business. New entries publish here when the thinking is ready, not when the calendar says. ## The Archive — 2026, Vol. 01 ## Anthropic's enterprise playbook, read from Muscat. Reading · 24 May 2026 · 8 min read Anthropic published a 35-page enterprise guide called Building trusted AI in the enterprise. We read it end-to-end. The short version: its four-stage spine maps almost exactly onto the engagement model Orfloat already sells. The longer version is what the playbook leaves out — and what a Muscat operator needs to add to make it land. [Read in full](https://www.orfloat.com/notes/anthropic-enterprise-playbook) ## AI is not a software business anymore. Briefing · 24 May 2026 · 9 min read Microsoft is spending $190 billion on capital expenditure in 2026 and still expects to run short on capacity. The four biggest hyperscalers will spend close to $700 billion combined this year — roughly 3.5× what they spent two years ago. Your AI vendor agreement is a supply contract in everything but name. [Read in full](https://www.orfloat.com/notes/ai-is-not-a-software-business) ## The capability overhang is no longer theoretical. Field note · 22 May 2026 · 6 min read Claude Opus 4.7 shipped six weeks ago. Anthropic's largest user study so far went out two months before that. The frontier is moving faster than most boards have agendas for — and almost none of that movement has reached operating businesses in the GCC. [Read in full](https://www.orfloat.com/notes/capability-overhang) ## Model Context Protocol, plainly explained. Primer · 19 May 2026 · 7 min read MCP is the boring-sounding standard that makes the rest of Anthropic's stack non-boring. If your business runs on a POS, an ERP, an inventory spreadsheet, and a calendar — this is the layer that finally lets Claude reason over all of them at once, without each integration becoming a special case. [Read in full](https://www.orfloat.com/notes/mcp-plainly-explained) ## Drafts — forthcoming The entries below are visible on the human-facing `/notes` page as "In draft" cards. They do not yet have published `/notes/` URLs — do not attempt to fetch them. They are listed here so the agent surface matches what human visitors see. ### What the typical 15-day Discovery actually looks like. **In draft · Method** A walkthrough of a typical on-site Discovery — who we sit with, what we measure, and what the deliverables read like on day fifteen. Publishing shortly. ### Capability overhang, Muscat edition (vol. 02). **In draft · Region** What we are learning from the first cohort of Omani operators who have stopped treating AI as a chatbot and started treating it as an operating layer. ## Begin **Every note began as a real problem. *Bring us yours.*** [Start a Discovery Phase](https://www.orfloat.com/contact) · [Why now](https://www.orfloat.com/why-now) --- # Forty-two agents on our own codebase, and the call they couldn't make > Source: https://www.orfloat.com/notes/dogfooding-the-workflow > Updated: 2026-05-29 # Forty-two agents on our own codebase, *and the call they couldn't make.* [Notes](https://www.orfloat.com/notes) · Field note · 29 May 2026 Yesterday we argued that the frontier has moved the binding constraint off the model and onto the disciplines around it — describing the work, and judging what comes back. It is one thing to write that. So the next morning we pointed Claude Opus 4.8 and Claude Code's new dynamic workflows at this website and let them run an end-to-end performance sprint, all the way to production. The workflow found a fix we should have caught months ago. It also nearly talked itself into shipping a mistake. Both halves are the point. ## The claim, and the test The note we published the day before — [Claude Opus 4.8, and the discipline it asks for](https://www.orfloat.com/notes/opus-4-8-discipline) — made a specific claim: a model that polices its own output, paired with orchestration that runs hundreds of cross-checking agents, stops being the bottleneck. What is left is the quality of the brief you hand it and the judgment you apply to what it returns. We did not want that to stay a thesis, so we ran it against the one codebase we are free to break: our own. The instrument was a [dynamic workflow](https://claude.com/blog/introducing-dynamic-workflows-in-claude-code) — a script Claude Code writes and a runtime executes in the background, orchestrating subagents at scale rather than working turn by turn. The brief was deliberately narrow: backend and platform performance only. Work on an isolated git worktree, never touch the main branch as the working tree, open a clean pull request, then merge. Freeze the design — no layout, no copy, no brand tone, no CSS tokens, and explicitly do not touch the *Reveal* scroll-animation observer this project has a documented history with. Pixel for pixel, the rendered site had to come out identical. The whole engagement lived or died on that one paragraph of constraints. That paragraph is Description — the first discipline — and it was the actual product of the morning. We set the session to *ultracode*, which pairs maximum reasoning effort with automatic workflow orchestration, and let it go. What came back, roughly fourteen minutes later, was a run that had spun up forty-two agents and spent about 1.57 million tokens — a six-auditor fan-out feeding a three-lens adversarial review, twelve candidate changes surfaced and cross-examined. ## What the run found Six changes survived review and shipped, in four small commits. The best of them was not in the application code at all — it was a single token in a response header. Our agent-readable routes (the `.md` twins, the `llms-*` files, the JSON-LD) were serving `Vary: Accept, Accept-Encoding, User-Agent`. Vercel's edge cache already keys on [Accept and Accept-Encoding by default](https://vercel.com/docs/caching/cdn-cache); the load-bearing mistake was the `User-Agent` token. It told the edge to store a separate cached copy for every distinct browser string, for responses that are byte-identical across all of them. As Vercel's own guidance puts it, each header you vary on multiplies the number of cache entries — and a multiplier keyed on User-Agent is effectively unbounded. Dropping it collapses thousands of needless variants back into one. It is the kind of bug that hides in a header for years because nothing is visibly broken; the site just runs colder at the edge than it should. The rest were quieter. The hero headline is set in our display serif, and its font preload was competing with the body-sans preload under bandwidth contention; the run added `fetchPriority="high"` to the serif alone, which [lifts it in the download queue](https://web.dev/articles/fetch-priority) so the [largest paint](https://web.dev/articles/vitals) is not gated on a font swap. The contact form's upstream send got a hard eight-second timeout and a guard around malformed submissions — both error-path only, the visible form untouched. And two genuinely dead components plus an unused icon were deleted: two hundred and eleven lines removed against twenty-two added. The bundle barely moved — client JavaScript went from 146,258 to 145,929 gzipped bytes — and that is the honest headline, not a disappointment. The dead code was already tree-shaken out of the shipped bundle, so deleting it cleaned the source without changing the payload. The entry chunk, the runtime, and the entire stylesheet kept byte-identical content hashes through the whole sprint. That invariance is the proof we actually wanted: the site got faster at the edge and in feel, and not one rendered byte changed to get there. ## What it talked itself out of Six of the twelve candidates were killed in review, and the discards are better evidence of the workflow's worth than the keeps. A fast pass would have shipped all twelve. This one argued itself out of the tempting-but-wrong ones, with specific reasons: - **Preloading the italic serif.** It would have put roughly 130KB on the critical path, contending with the actual largest-paint font. A speed change that costs speed. - **Stripping the loader headers.** Rejected on a false premise: a prior commit in our own history documents that AI crawlers need those headers to accept the response. The workflow read the paper trail and stood down. - **Switching the build target to a newer baseline.** A no-op — the bundler already emits modern output. No change, no merge. - **Sharing one Reveal observer across the page.** High risk, low reward, on the exact file our project memory flags as having a non-obvious correctness constraint. Left alone. That is taste, and taste is what makes more agents worth more rather than just louder. The value was never the raw output. It was the adversarial layer deciding what not to keep. ## Where it nearly fooled itself And then it nearly shipped the wrong call anyway. Two of the audit agents, told explicitly to read and not write, edited files on disk. One of those stray edits then poisoned a downstream verdict: a verifier agent opened an already-mutated file, found the icon it was asked to check apparently still in use, and concluded the dead code was not dead. A clean majority of agents, reasoning over a contaminated working tree, was about to vote the right deletion off the list. > A confident agent reading a corrupted file is indistinguishable, on the surface, from a correct one. What caught it was not another vote. It was the orchestrator noticing that an agent's verdict contradicted the ground truth — a plain search of the repository — resetting to a clean baseline, re-applying every change by hand so the final diff was fully intentional, and then independently re-confirming the code really was dead before deleting it. That is the second and third disciplines doing exactly the job we said they would have to: Discernment, reading the agent's decision rather than trusting its confidence; and Diligence, refusing to act on a "clean tree" nobody had verified. The lesson generalizes, and it is the one piece of this we would put in front of the people building these tools. Audit agents need a hard read-only boundary the harness enforces, not a polite instruction in a prompt. And majority agreement across agents is not truth — when their inputs can be silently corrupted by a sibling, a vote only launders the error. An agent's verdict has to be treated as advice, checked against the world, never as a fact. The orchestrator that does that checking is not overhead. On this run it was the only thing standing between a good sprint and a quietly broken one. ## What to do with this The sprint is live. It merged in one commit, it is reversible in one command, the design is pixel-for-pixel unchanged, and the site feels measurably snappier on both phones and desktops in production. We are being deliberately careful with that last claim: the real gains are at Vercel's edge, where they do not show up on a local machine, so the defensible numbers are field data we are still collecting — not a benchmark we ran on a loopback and dressed up. The mechanism is sound and the feel is real; the figures will follow, honestly or not at all. For an operator in Muscat watching this from the outside, the transferable part is not "use the new model." It is the shape of the day. A capable workflow did hours of audit labour in minutes and surfaced a fix worth real money at scale. It also made a mistake that no amount of model quality would have caught on its own. The difference between those two outcomes was a human-held brief and a human-held review — the disciplines, not the model. That is the whole of what we do, and we just watched it hold up on our own codebase before we asked anyone to trust it with theirs. If you want a sprint like this run inside your business — scoped, reversible, and judged by someone who reads the diff — [start a conversation](https://www.orfloat.com/contact). — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Anthropic. *Orchestrate subagents at scale with dynamic workflows.* 28 May 2026. [claude.com/blog/introducing-dynamic-workflows-in-claude-code](https://claude.com/blog/introducing-dynamic-workflows-in-claude-code) 2. Vercel. *Vercel CDN Cache.* 2026. [vercel.com/docs/caching/cdn-cache](https://vercel.com/docs/caching/cdn-cache) 3. web.dev. *Optimize resource loading with the Fetch Priority API.* [web.dev/articles/fetch-priority](https://web.dev/articles/fetch-priority) 4. web.dev. *Web Vitals.* [web.dev/articles/vitals](https://web.dev/articles/vitals) 5. Orfloat. *Claude Opus 4.8, and the discipline it asks for.* 28 May 2026. [orfloat.com/notes/opus-4-8-discipline](https://www.orfloat.com/notes/opus-4-8-discipline) [← Back to Notes](https://www.orfloat.com/notes) --- # Claude Opus 4.8, and the discipline it asks for > Source: https://www.orfloat.com/notes/opus-4-8-discipline > Updated: 2026-05-28 # Claude Opus 4.8, *and the discipline it asks for.* [Notes](https://www.orfloat.com/notes) · Field note · 28 May 2026 Anthropic released Claude Opus 4.8 today, and the headline is not a benchmark. The model is around four times less likely than its predecessor to let flaws in its own code pass unremarked — it catches its own mistakes and pushes back when a plan is wrong. In the same release, Claude Code got dynamic workflows: one session can now orchestrate hundreds of subagents that review each other's work before anything reaches you. Put those two together and the binding constraint on an operating business has moved somewhere most boards are not looking. ## What actually shipped The benchmark line moved, as it always does. On agentic coding (SWE-Bench Pro) Opus 4.8 [scores 69.2%](https://officechai.com/ai/claude-opus-4-8-benchmarks/), up from 64.3% for Opus 4.7 and well ahead of the other frontier models reported at 58.6% and 54.2%. On agentic computer use (OSWorld-Verified) it reaches 83.4%, and on browser-agent tasks (Online-Mind2Web) Anthropic reports [84% — the strongest it has tested](https://www.anthropic.com/news/claude-opus-4-8), along with being the first model to complete every case end-to-end on its Super-Agent benchmark. Pricing is unchanged at $5 per million input tokens and $25 per million output, and fast mode is now three times cheaper than it was on previous models. None of that is the part worth reorganizing around. The part worth reorganizing around is the change in honesty. Anthropic trains its models not to make claims they cannot support, and 4.8 is the first release where that training shows up as a number an engineering manager can feel: it is [roughly four times less likely](https://www.tomsguide.com/ai/claude-opus-4-8-just-launched-and-anthropic-says-its-far-less-likely-to-fake-answers) than Opus 4.7 to let a flaw in its own code pass without flagging it. Anthropic's alignment team puts its rates of misaligned behaviour — deception, cooperation with misuse — substantially below 4.7. For anyone who has spent a year reading agent output with one eyebrow raised, a model that is measurably more willing to say "I am not sure this is right" is the upgrade that matters. ## Orchestration stops being a metaphor The second half of the release is in Claude Code. [Dynamic workflows](https://claude.com/blog/introducing-dynamic-workflows-in-claude-code) let Claude write a script that orchestrates subagents at scale — a runtime executes it in the background while your session stays responsive. The constraints are concrete: up to sixteen agents run concurrently, up to a thousand across a single run. You reach for one when a task needs more agents than a single conversation can coordinate — a codebase-wide bug sweep, a five-hundred-file migration, a research question whose sources need cross-checking against each other. The mechanism is more interesting than the scale. With subagents and skills, Claude is the orchestrator: it decides turn by turn what to spawn next, and every intermediate result lands back in its context. A workflow moves the plan into code. The script holds the loop, the branching, and the intermediate results, so Claude's context holds only the final answer. That is what lets a workflow apply a repeatable quality pattern rather than just running more agents — it can have independent agents *adversarially review* each other's findings before they are reported, or draft a plan from several angles and weigh them against one another. > Work you would normally plan in quarters now finishes in days. That line is Anthropic's, and it is the kind of claim we normally discount. The reason to take this one seriously is that the orchestration is now legible: the workflow is a script you can read, save as a command, and rerun on every branch. A review that fans out sixteen agents to cross-examine a diff, votes on what survives, and hands you a single cited verdict is not a demo — it is a process you can own. Anthropic also shipped *ultracode*, a setting that lets Claude decide on its own when a task warrants a workflow, so the orchestration becomes the default rather than something you invoke by hand. ## The constraint moved again We wrote [the capability overhang](https://www.orfloat.com/notes/capability-overhang) — the gap between what frontier models can already do and what operating businesses actually use them for — as the spine of an earlier note. Each release widens it. But 4.8 widens it in a specific direction. The two features that shipped today attack the two oldest reasons an operator gave for not delegating real work to an agent: you cannot trust the output, and one agent cannot hold a job big enough to matter. A model that polices its own code and a runtime that runs hundreds of cross-checking agents answer both at once. So the binding constraint is no longer the model. It is not compute, and for most GCC businesses it was never the budget. The constraint is the quality of the instruction the agent is given and the quality of the judgment applied to what it returns. When the hard part of software was writing correct code, capability was the bottleneck. Now that an orchestrated run can produce, cross-check, and verify the code, the bottleneck is upstream and downstream of the model — in the brief and in the review. That is not a technology problem. It is an operating-discipline problem, and it does not get solved by buying a larger model. ## Which discipline it tests Inside the studio we work from four operating disciplines — the 4D framework. Today's release does not touch all four evenly. It presses hardest on the middle two, and it quietly changes the shape of the fourth. - **Delegation.** Still the entry point — decide what a workflow owns and where it stops. With a thousand-agent ceiling, the question is no longer "can I delegate this" but "what is the unit I am delegating." - **Description.** The binding input. A workflow executes the brief you gave it across hundreds of agents; a vague brief now fails at scale, not in one conversation. The model being more honest does not rescue a description that never said what "done" meant. - **Discernment.** The binding output. When a run returns one cited verdict instead of a turn-by-turn transcript, your job is to read the judgment, not the keystrokes — did it pick the right thing, in the right order, with the right tradeoffs. This is the skill that compounds. - **Diligence.** Changed in shape. You no longer audit each line; you audit the orchestration — what the agents could touch, what the cross-check actually checked, where the run could go wrong unobserved. Re-audit on the old cadence, because the ceiling moved again today. The honesty improvement is real, and it helps. But it is a floor, not a ceiling. A model that flags its own uncertainty more often still needs someone whose judgment is good enough to know which flags matter — and a brief precise enough that the flags are about the work, not about what the work was supposed to be. ## What to do with this If you run an operating business in Muscat or the wider GCC, the move this release argues for is not "adopt 4.8." The model will reach you whether you plan for it or not. The move is to build the two disciplines the model now demands: a small team that can write a brief precise enough to survive a hundred-agent run, and read the result with enough judgment to sign it. That is a capability you grow inside your own operation, against your own data and your own consequences — it is not a licence you buy. That team — judgment-heavy, close to the real systems, building the description-and-discernment muscle while the capability curve is still ahead of habit — is exactly what a Discovery Phase is. If you would like one inside your business, [start a conversation](https://www.orfloat.com/contact). — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Anthropic. *Introducing Claude Opus 4.8.* 28 May 2026. [anthropic.com/news/claude-opus-4-8](https://www.anthropic.com/news/claude-opus-4-8) 2. Anthropic. *Orchestrate subagents at scale with dynamic workflows.* 28 May 2026. [claude.com/blog/introducing-dynamic-workflows-in-claude-code](https://claude.com/blog/introducing-dynamic-workflows-in-claude-code) 3. OfficeChai. *Anthropic Releases Claude Opus 4.8, Beats Opus 4.7, GPT-5.5 On Many Benchmarks.* 28 May 2026. [officechai.com/ai/claude-opus-4-8-benchmarks](https://officechai.com/ai/claude-opus-4-8-benchmarks/) 4. VentureBeat. *Anthropic's Claude Opus 4.8 is here with 3X cheaper fast mode and near-Mythos level alignment.* 28 May 2026. [venturebeat.com/technology](https://venturebeat.com/technology/anthropics-claude-opus-4-8-is-here-with-3x-cheaper-fast-mode-and-near-mythos-level-alignment) 5. Tom's Guide. *Claude Opus 4.8 just launched — and Anthropic says it's far less likely to 'fake' answers.* 28 May 2026. [tomsguide.com/ai](https://www.tomsguide.com/ai/claude-opus-4-8-just-launched-and-anthropic-says-its-far-less-likely-to-fake-answers) 6. Techzine. *Anthropic releases Claude Opus 4.8, promising a more honest model.* 28 May 2026. [techzine.eu/news/applications](https://www.techzine.eu/news/applications/141667/anthropic-releases-claude-opus-4-8-promising-a-more-honest-model/) [← Back to Notes](https://www.orfloat.com/notes) --- # Software after software, and the record so far > Source: https://www.orfloat.com/notes/software-after-software > Updated: 2026-05-27 # Software after software, *and the record so far.* [Notes](https://www.orfloat.com/notes) · Field note · 27 May 2026 "Software After Software" is a short manifesto we have been working from inside the studio for the last few months. Twelve numbered propositions, in the Tractatus tradition, on what software becomes when intelligence is abundant, continuous, and cheap. We had treated it as a working thesis until the last six months. The empirical record has now overtaken it. The manifesto is not speculative anymore. Every load-bearing claim has a corresponding line in the data — and the operators who treat this as future-of-work reading are already a quarter behind. ## The capability curve, refreshed [The capability overhang](https://www.orfloat.com/notes/capability-overhang) — the gap between what frontier models can already do and what operating businesses are actually using them for — was the spine of an earlier note in this archive. Three things have moved in the six months since. First, agent autonomy is now measurable, and it is climbing. In a research piece published this year, [Anthropic reported](https://www.anthropic.com/research/measuring-agent-autonomy) that the 99.9th-percentile turn duration for Claude — the elapsed time between an agent starting work and stopping — nearly doubled in roughly three months, from under twenty-five minutes in late September 2025 to over forty-five minutes by early January 2026. The language Anthropic chose for the piece is itself striking: there is a *deployment overhang* — the autonomy the models are capable of handling exceeds what they exercise in practice. Capability is now running ahead of habit, not the other way around. Second, the practitioners with the largest individual footprint on Claude have made the shift in name as well as in practice. In February 2026, Andrej Karpathy — who coined "vibe coding" exactly one year earlier — declared the term effectively over. The new default, in his framing, is *agentic engineering*: you are not writing the code directly 99% of the time, you are orchestrating agents who do and acting as oversight. The vocabulary has moved from convenience to discipline. Third, the supply side has put very large numbers behind all of this. The four largest hyperscalers — Microsoft, Google, Amazon, Meta — collectively plan [$725 billion in capital expenditure in 2026](https://www.tomshardware.com/tech-industry/big-tech/big-techs-ai-spending-plans-reach-725-billion), up 77% from last year's record $410 billion. Microsoft alone is at $190 billion, well above the $152 billion analysts had been modelling. The composition has shifted too: more than 60% of that spend is now going to power infrastructure rather than chips — a structural change in where the binding constraint actually sits. ## The bottleneck moved The manifesto's third and fourth theses are about where the constraint sits inside a software organization. The summary: writing valid code is trivial now. What remains are errors of engineering — priorities, sequencing, tradeoffs — and these are the errors that matter. > You are not writing the code directly 99% of the time. You are orchestrating agents who do. Karpathy's framing maps almost word-for-word onto the manifesto's claim that the unit of work becomes the delegated task, not the code to be written. The implication for review is concrete: review shifts from code to decisions. The pull request you read is no longer "is this implementation correct" — that question is now settled before it reaches you. The pull request you read is "did the agent pick the right thing to build, in the right order, with the right tradeoffs against the cost of building the other thing." That is a different skill, and it is the only skill that compounds from here. ## The old assumptions break Two of the manifesto's harder claims: software, as a *profession*, was built on the assumption that writing code is hard and error-prone. Software, as an *industry*, was built on the assumption that code is scarce. Both assumptions no longer hold. When code stops being scarce, the value migration is mechanical. Software whose only job is to encode a workflow loses value the moment an agent can perform the workflow directly. The moat for thousands of mid-market SaaS vendors — "customers cannot justify building this themselves" — becomes a clearance sale. What gains value, in this picture, is everything an agent cannot produce on demand: proprietary data, distribution and customer relationships, regulatory position, physical assets, trust, and the permissions to operate in a regulated space. The hyperscaler capex composition is the same migration in a different register. The line item that grew fastest in 2026 was not chips — it was power, and the physical infrastructure that delivers it. Compute is being commoditized as fast as the supply can be built; what remains scarce is the substrate underneath it. ## Organize around the models The manifesto's ninth and tenth theses are the ones that hit operating businesses hardest. It is not enough to fit models into existing systems, org charts, and processes. The winners are those who organize *around* the models. A small team with strong judgment and many agents will outrun a large team trying to fit AI into processes designed before the transformation. Inside the studio, our four operating disciplines apply to this directly. We call them the 4D framework, and they map cleanly onto the manifesto's organizational claim: - **Delegation.** Decide what each agent owns, and what it does not. An agent forced to work like a human is a wasted agent — the manifesto says it; our engagements measure it. - **Description.** Tell the agent how to use the tools at its disposal, in writing, the way you would brief a thoughtful new colleague. The model's behaviour is your specification. - **Discernment.** Read the agent's decisions the way you would read a junior engineer's pull request — for taste, sequencing, tradeoff selection. The code is incidental. - **Diligence.** Audit the surface area before granting access. Re-audit every eight weeks, because the capability curve will have moved. For a GCC operator, none of this is abstract. A family business in Muscat that is running its hospitality group, its clinic, or its retail operation through a layer of agents in 2027 — and many will — must have started reorganizing around the models in 2026. Bolting AI onto the old way of working is a category error, not an integration project. ## What to do with this The manifesto closes with five short "Therefore" lines. The one that is hardest to argue with: we do not wait for the end state to become obvious, because the best move is to play the game. The end state is not yet apparent — but the curve is, and the curve is now enough to act on. For an Omani business in May 2026, the concrete move is to seat a small team — three or four people, judgment-heavy, low on process — close enough to your real systems, real data, and real consequences to discover the new way of working inside your specific operation. That team's job is not to add AI to anything. Its job is to find the new shape of the work and pull the rest of the organization toward it. Its output, as the manifesto says, is not only software but also people and practices. That kind of small autonomous team is exactly what a Discovery Phase is. If you would like one inside your business, [start a conversation](https://www.orfloat.com/contact). — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Anthropic. *Measuring agent autonomy.* 2026. [anthropic.com/research/measuring-agent-autonomy](https://www.anthropic.com/research/measuring-agent-autonomy) 2. The New Stack. *Vibe coding is passé. Karpathy has a new name for the future of software.* 2026. [thenewstack.io/vibe-coding-is-passe](https://thenewstack.io/vibe-coding-is-passe/) 3. Tom's Hardware. *Google, Microsoft, Meta, and Amazon capex spending to hit $725 billion in 2026, up 77% from last year.* 2026. [tomshardware.com/tech-industry/big-tech/big-techs-ai-spending-plans-reach-725-billion](https://www.tomshardware.com/tech-industry/big-tech/big-techs-ai-spending-plans-reach-725-billion) 4. CNBC. *Tech AI spending approaches $700 billion in 2026, cash taking big hit.* 6 February 2026. [cnbc.com/2026/02/06/google-microsoft-meta-amazon-ai-cash.html](https://www.cnbc.com/2026/02/06/google-microsoft-meta-amazon-ai-cash.html) 5. Orfloat. *The capability overhang is no longer theoretical.* 22 May 2026. [orfloat.com/notes/capability-overhang](https://www.orfloat.com/notes/capability-overhang) [← Back to Notes](https://www.orfloat.com/notes) --- # Anthropic's enterprise playbook, read from Muscat > Source: https://www.orfloat.com/notes/anthropic-enterprise-playbook > Updated: 2026-05-24 # Anthropic's enterprise playbook, *read from Muscat.* [Notes](https://www.orfloat.com/notes) · Reading · 24 May 2026 Anthropic published a 35-page enterprise guide called *Building trusted AI in the enterprise*. We read it end-to-end this week. The short version: it describes a four-stage spine that almost exactly matches the engagement model we already sell, and one we now have explicit language for. The longer version is what the playbook leaves out — what a Muscat operator needs to add to make it land. You can read the original guide in full on [anthropic.com](https://www.anthropic.com/news) — we won't reproduce it here. What follows is our commentary as forward-deployed engineers operating in the Gulf: what we adopt from it as written, where it has to be translated to land in a family-led GCC business, and where Orfloat's own thinking goes further than the guide. ## The four-stage spine The guide organises an enterprise AI programme into four sequential stages: develop a strategy, create business value through a pilot, build for production, and then deploy with LLMOps. Anthropic notes that companies with the right motivation can compress what they otherwise frame as a 13-month rollout into a few months — citing FeatherSnap integrating Claude on Amazon Bedrock in under 90 days and DoorDash building a voice contact-centre solution in two months. We are familiar with that timeline. Our typical engagement runs about 16 weeks from on-site Discovery to first production system. The four-stage model maps cleanly onto our own: - **Stage 1 (Strategy) ↔ Orfloat Discovery.** Their guide says to start with people, process, and technology. Our 15-day on-site Discovery does exactly that — three of the deliverables (governance map, opportunity priority list, technical-readiness audit) are Anthropic's three dimensions, named differently. - **Stage 2 (Business value) ↔ Service Agreement.** Anthropic's seven-criterion pilot test — LLM-suited work, measurable metrics, clear ROI, business-critical but low security risk, abundant data, minimal disruption, scalable — is the same checklist we apply when we draft the milestone-based scope after Discovery. - **Stage 3 (Production) ↔ Forward deployment, weeks 1–8.** The prompt-engineering structure they spell out (task+role, background, rules, history, request, format, prefill) is our default scaffold. Their emphasis on evaluation-before-deployment is non-negotiable for us. - **Stage 4 (LLMOps) ↔ Forward deployment, weeks 9–12 + 90-day handover.** Their five LLMOps practices — monitoring, prompt version control, security by design, scalable infrastructure, continuous QA — are the operating discipline we install before we leave. ## People, Process, Technology — translated for a Muscat family business The guide's three-dimensional model is correct, and it is also written for a different kind of company than the ones we work with. Three places where translation is necessary: > The guide assumes you have an executive sponsor, a steering committee, and a head of AI. Most Omani family businesses have a managing director, his cousin who runs operations, and an IT vendor. **People.** Anthropic prescribes "executive alignment and sponsorship" and an "AI review board." In the GCC family-business context, that is the founder and the COO sitting in the same Discovery workshop, and the AI review board is the same group that already meets weekly to talk supplier prices. We don't create new committees. We meet the existing ones where they already convene, and we hand them better questions. **Process.** Anthropic's pilot-graduation criteria — performance thresholds, operational readiness, risk management infrastructure — are the right gates. The guide implicitly assumes you have a product analytics practice in place to measure those gates. Most of our clients don't. Part of our forward-deployed work is instrumenting the operation enough that the criteria can be measured at all. The pilot doesn't fail; the measurement framework fails first. **Technology.** The guide describes a clean three-level technical maturity: basic chat, intermediate with RAG and tools, advanced agents. In a Muscat operation, you may need to be at all three levels simultaneously — a basic chatbot for guests, a RAG-equipped concierge for returning customers, and an autonomous nightly reconciliation agent for the back office. The progression in the guide is accurate at the enterprise level; at the individual-business level, you pick the right level for the right workflow. ## The 12-month clock, in practice Anthropic's four-phase rollout — Foundation (months 1–3), Pilot (4–6), Strategic Scaling (7–12), Broad Adoption (13+) — is the right cadence for a large enterprise. For a GCC family business with 80–400 staff, we compress it considerably: Foundation in weeks 1–3, Pilot in weeks 4–10, first production system live by week 12, second production system by week 16. The four-phase logic is preserved. Only the calendar shrinks. Anthropic's own note that "motivation and partnership" can condense the timeline to weeks is exactly the lever we pull. ## The LLMOps gap is the engagement Anthropic cites a BCG survey of 1,400 C-suite executives in which 62% identified shortage of talent and skills as the biggest obstacle to their AI strategy. Inside an Omani operator's building, that number is closer to 100% — not because the talent doesn't exist, but because no single hire fills the shape of the role. The shape is one part LLM engineer, one part operations specialist, one part data architect, one part compliance reader. Hiring that profile in Muscat in 2026 is not a recruitment problem; it is a scarcity problem. Forward deployment exists to fill the shape without making the client own it. ## What we adopt, and what we add We adopt the playbook's spine — the four stages, the three dimensions, the seven pilot criteria, the five LLMOps practices — directly. We treat it as the published standard for serious enterprise AI work. Where we add: a translation layer for the GCC family business, a measurement-instrumentation layer that the playbook assumes you already have, a Muscat-aware governance cadence, and a forward-deployed delivery model that compresses the calendar. If you are running an operating business in the Gulf and the four-stage model above sounds like a useful map — the right next step is to find out what your Stage 1 looks like honestly. [Start a Discovery Phase.](https://www.orfloat.com/contact) — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Anthropic. *Building trusted AI in the enterprise — Anthropic's guide to starting, scaling, and succeeding based on real-world examples and best practices.* See [anthropic.com/news](https://www.anthropic.com/news) for the latest enterprise resources. Trademark and attribution acknowledged on our [Trademarks page](https://www.orfloat.com/trademarks). 2. Bain & Company. *Technology Report 2024.* [bain.com](https://www.bain.com/insights/topics/technology-report/) 3. McKinsey & Company. *The state of AI in 2024.* [mckinsey.com](https://www.mckinsey.com/capabilities/quantumblack/our-insights/the-state-of-ai) 4. BCG. *Five Must-Haves for Effective AI Upskilling.* [bcg.com](https://www.bcg.com/publications/2024/five-must-haves-for-effective-ai-upskilling) 5. Anthropic. *Prompt engineering documentation.* [docs.claude.com](https://docs.claude.com) [← Back to Notes](https://www.orfloat.com/notes) --- # AI is not a software business anymore > Source: https://www.orfloat.com/notes/ai-is-not-a-software-business > Updated: 2026-05-24 # AI is not a *software business* anymore. [Notes](https://www.orfloat.com/notes) · Briefing · 24 May 2026 Microsoft will spend roughly **$190 billion** on capital expenditure in calendar 2026 — and still expects to be capacity-constrained through year-end. The four biggest hyperscalers will spend close to **$700 billion** combined this year, roughly 3.5× what they spent two years ago. None of that looks like a classic software business. Six months from now, neither will your AI vendor contract. The cloud was supposed to make infrastructure someone else's problem. Write code once, run it many times, scale on demand. The whole abstraction depended on supply staying ahead of demand by a comfortable margin. AI broke that abstraction — not gradually, and not theoretically. The Q3 FY26 results from Microsoft, published in late April, put the new shape of the industry on the public record. ## The numbers tell you what changed Microsoft reported [$31.9 billion in capital expenditure](https://www.sec.gov/Archives/edgar/data/0000789019/000119312526191507/msft-20260331.htm) for the fiscal third quarter ending 31 March 2026, and guided to **more than $40 billion** for the following quarter. About two-thirds of the quarterly spend went to short-lived assets — primarily GPUs and CPUs that will be fully depreciated inside four years. For calendar 2026, the company expects total capex to land around [$190 billion](https://www.cnbc.com/2026/04/29/microsoft-msft-q3-earnings-report-2026.html) — and the CFO was explicit that this will not be enough. Azure demand still exceeds supply. Microsoft is not alone. Alphabet, Amazon, and Meta are running the same play. Combined hyperscaler 2026 capex is now [projected to approach $700 billion](https://fortune.com/2026/04/30/big-tech-hyperscalers-will-spend-700-billion-on-ai-infrastructure-this-year-with-no-clear-end-in-sight-eye-on-ai/) — up from roughly $200 billion in 2024 and 6× the 2022 level. Meta raised its full-year guidance to $125–145 billion and explicitly blamed component and data-center costs. Amazon is now expected to post a [negative free cash flow year](https://www.cnbc.com/2026/02/06/google-microsoft-meta-amazon-ai-cash.html) for the first time in over a decade. Mary Meeker, in her [340-page Bond Capital report](https://www.bondcap.com/reports/tai) published last May, used the word "unprecedented" on 51 separate pages to describe the rate of change. At NVIDIA's GTC 2026 keynote, Jensen Huang stopped calling Nvidia a chip company at all. He calls it an [AI factory company](https://www.datacenterfrontier.com/machine-learning/news/55364406/jensen-huang-maps-the-ai-factory-era-at-nvidia-gtc-2026) now. "Tokens are the new commodity," he said twice on the same slide. He was not being poetic. ## Tokens are not magic. They are manufactured. The most important shift to understand is that every answer from a model is the output of a physical production system. GPUs, high-bandwidth memory, advanced packaging, substrates, optics, power, cooling, land, networking, and operations talent — that is the bill of materials behind the paragraph of text on your screen, the line of code Claude wrote you, or the agent that just finished summarising a contract. The user sees software. Behind the software is a factory turning electricity and silicon into intelligence. > Six months ago an AI vendor agreement was structured like a software agreement. Today it is a supply contract in everything but name. And that supply contract has terms most procurement teams have never written before: allocation, fallback, reserved capacity, multi-region failover. None of it was a line item a year ago. All of it should be in the next one you sign. ## What this changes for an operator in Muscat At first glance, hyperscaler capex looks like a Silicon Valley story. It is not. The Gulf has been positioning itself as a sovereign AI-infrastructure region for a year and a half, and the placements are now public. Saudi Arabia's [HUMAIN](https://www.globaldatacenterhub.com/p/does-humains-12b-saudi-framework) secured a $1.2 billion AI-infrastructure framework in January. The UAE's G42, partnered with Oracle, NVIDIA, Cisco, and SoftBank under the [Stargate UAE](https://introl.com/blog/middle-east-ai-revolution-uae-saudi-arabia-100b-infrastructure-plans) banner, is building a 5 GW AI campus in Abu Dhabi with a 1 GW cluster going live in 2026. Oman's own *AI and Digital Future Programme* names data-centre capacity as a delivery mechanism for the digital economy. For an operating business in Muscat, this means two things at once. The first is that *closer* inference is finally becoming a procurement option — your token round-trip will be in-region within the next eighteen months, with the residency and latency profile that regulated industries need. The second is that the same capacity rationing playing out in Redmond and Mountain View will play out in Riyadh and Abu Dhabi too, just on a slight lag. Whoever signs first sits at the front of the queue. ## The unit that actually matters is the token Almost every AI plan we read inside client engagements is still priced in seats. Five Claude Pro seats. Ten Copilot seats. A team licence. Seats are the wrong unit for a factory output. The right unit is the token — input tokens, output tokens, cached tokens, and the routing decisions between them. We forecast client workloads in tokens. We treat an agent that runs unattended for an hour as a different budget line than a chatbot that answers one question. We instrument every workflow to know — by name — which model was called, with what prompt, returning how many tokens, at what cost. None of that is exotic. It is just bookkeeping that nobody has bothered to install yet, because the abstraction of "seats" let everyone pretend the supply was infinite. It isn't. ## What we do about it Inside a Discovery Phase, the supply-chain reality of AI shows up in three concrete deliverables. - **A vendor audit.** We map the supply chain under your current AI contracts — which cloud, which region, which underlying chip family, what allocation terms exist (or don't), and what your fallback looks like when the next capacity squeeze hits. - **A token-denominated forecast.** We size demand in tokens, not seats — separating chatbot work from agentic workloads and pricing each at the model class that actually fits. - **A routing diagnostic.** Most operations burn premium-tier inference (Opus-class) on work that Haiku-class would do for a tenth of the cost. We find those, fix them, and the savings usually pay for the engagement before deployment. Microsoft has already put $190 billion behind a view of the world. Most operating businesses in the Gulf have not put even the equivalent procurement framework into their plans. That is the gap. It will close — for the businesses that decide to close it. [Start a Discovery Phase.](https://www.orfloat.com/contact) — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Microsoft Corp. *Form 10-Q, Fiscal Q3 2026 (quarter ended 31 March 2026).* [sec.gov/Archives/edgar/data/0000789019/…/msft-20260331](https://www.sec.gov/Archives/edgar/data/0000789019/000119312526191507/msft-20260331.htm) 2. CNBC. *Microsoft calls for $190 billion in 2026 capital spending on soaring memory prices.* 29 April 2026. [cnbc.com](https://www.cnbc.com/2026/04/29/microsoft-msft-q3-earnings-report-2026.html) 3. Fortune. *Big Tech is about to spend $700 billion on AI this year.* 30 April 2026. [fortune.com](https://fortune.com/2026/04/30/big-tech-hyperscalers-will-spend-700-billion-on-ai-infrastructure-this-year-with-no-clear-end-in-sight-eye-on-ai/) 4. Bond Capital / Mary Meeker. *Trends — Artificial Intelligence.* May 2025. [bondcap.com/reports/tai](https://www.bondcap.com/reports/tai) 5. Data Center Frontier. *Jensen Huang Maps the AI Factory Era at NVIDIA GTC 2026.* March 2026. [datacenterfrontier.com](https://www.datacenterfrontier.com/machine-learning/news/55364406/jensen-huang-maps-the-ai-factory-era-at-nvidia-gtc-2026) 6. Global Data Center Hub. *Does HUMAIN's $1.2B Saudi Framework Signal a New Model for AI Data Centers?* January 2026. [globaldatacenterhub.com](https://www.globaldatacenterhub.com/p/does-humains-12b-saudi-framework) 7. Introl. *Middle East AI Revolution: UAE and Saudi Arabia's $100B+ Infrastructure Plans.* [introl.com](https://introl.com/blog/middle-east-ai-revolution-uae-saudi-arabia-100b-infrastructure-plans) [← Back to Notes](https://www.orfloat.com/notes) --- # The capability overhang is no longer theoretical > Source: https://www.orfloat.com/notes/capability-overhang > Updated: 2026-05-22 # The capability overhang is *no longer theoretical.* [Notes](https://www.orfloat.com/notes) · Field note · 22 May 2026 Six weeks ago Anthropic shipped Claude Opus 4.7. Two months before that, they published the largest multilingual study of AI users ever attempted. The frontier is moving faster than most boards have agendas for — and almost none of that movement has reached operating businesses in the Gulf. The gap between what a 2026 frontier model can do and what most businesses are doing with AI is now wide enough that it is no longer a forecasting question. It is a planning question. ## The frontier moved. Did you? In April, Anthropic released [Claude Opus 4.7](https://www.anthropic.com/news/claude-opus-4-7) — an upgrade that, in Anthropic's own framing, raised the bar on coding, agents, vision, and multi-step reasoning. The same week, Anthropic Labs launched [Claude Design](https://www.anthropic.com/news/claude-design-anthropic-labs) — a tool that does in fifteen minutes what an agency would charge a small business in Muscat OMR 800 for last year. Neither of these announcements is news to the AI-curious. Neither has meaningfully changed how a hospitality group in Al Mouj or a clinic in Qurum operates this Wednesday. In March, Anthropic published [*What 81,000 people want from AI*](https://www.anthropic.com/81k-interviews) — the largest multilingual qualitative study of AI users ever run. The most striking finding, for our purposes, is not what people are using AI for. It is the gulf between what power-users get out of Claude and what casual users do. The same tool, in the same week, delivered different orders of magnitude of value to different people. That gap is the capability overhang. ## The MENA picture, briefly PwC, in their since-foundational [*The potential impact of AI in the Middle East*](https://www.pwc.com/m1/en/publications/potential-impact-artificial-intelligence-middle-east.html) report, projected that AI would contribute roughly **USD 320 billion to the Middle East economy by 2030** — with the UAE and Saudi Arabia capturing the lion's share and Oman, Bahrain, and Kuwait splitting a long tail. The Oman government's own [Vision 2040](https://www.oman2040.om) names a knowledge economy and digital transformation as a first-class priority. The framing is there. The execution cadence inside privately-held businesses is not. What we see, walking into a small or mid-size enterprise in Muscat today, is not absence of intent. It is absence of a method that survives contact with a normal Tuesday. Owners read about Claude. Operations managers try ChatGPT. The integration with the actual POS, the actual supplier WhatsApp thread, the actual reservations system, never happens — because there is no one inside the business whose job it is to make it happen, and the agencies they hire to build websites do not do this work. ## Why this is not a tooling problem You can read the entire Anthropic blog this weekend. You can run Claude Code locally. You can buy Claude Pro on personal credit cards across five staff. None of those moves close the overhang. Because the overhang is not knowledge. It is embedding. > The gap between what AI *can already do* and what most teams are actually using it for is now wider than the gap between proprietary models and open-source ones — and changes faster. Forward-deployed engineering, as the practice has been described inside Anthropic and [Stripe](https://stripe.com/blog/forward-deployed-engineers) and [Palantir](https://www.palantir.com/offerings/foundry/) before them, is the answer to this. Not consulting. Not training decks. Engineers who sit inside the business, learn the work end to end, and ship. ## What we are doing about it Orfloat is one studio, in one city, taking on a small number of engagements at a time. We are not large enough to close the regional overhang. We are large enough to close it for a few businesses we choose carefully. If you have read this far and the gap I am describing sounds familiar — start a [Discovery Phase](https://www.orfloat.com/contact) and let's look at your operation honestly. — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Anthropic. *Introducing Claude Opus 4.7.* 16 April 2026. [anthropic.com/news/claude-opus-4-7](https://www.anthropic.com/news/claude-opus-4-7) 2. Anthropic. *Introducing Claude Design by Anthropic Labs.* 17 April 2026. [anthropic.com/news/claude-design-anthropic-labs](https://www.anthropic.com/news/claude-design-anthropic-labs) 3. Anthropic. *What 81,000 people want from AI.* 18 March 2026. [anthropic.com/81k-interviews](https://www.anthropic.com/81k-interviews) 4. PwC Middle East. *The potential impact of Artificial Intelligence in the Middle East.* [pwc.com/m1](https://www.pwc.com/m1/en/publications/potential-impact-artificial-intelligence-middle-east.html) 5. Sultanate of Oman. *Oman Vision 2040.* [oman2040.om](https://www.oman2040.om) [← Back to Notes](https://www.orfloat.com/notes) --- # Model Context Protocol, plainly explained > Source: https://www.orfloat.com/notes/mcp-plainly-explained > Updated: 2026-05-19 # Model Context Protocol, *plainly explained.* [Notes](https://www.orfloat.com/notes) · Primer · 19 May 2026 MCP is the boring-sounding standard that makes the rest of Anthropic's stack non-boring. If your business runs on a POS, an ERP, an inventory spreadsheet, and a calendar — this is the layer that finally lets Claude reason over all of them at once, without each integration becoming a special case. ## The problem MCP solves A large language model on its own is a closed system. It knows what it knows from training, and it knows what you tell it in the current chat. The moment you ask it to do something useful for your business — "is this guest already a regular?", "reorder the saffron when stock drops below 200g", "what is on the kitchen prep list for tomorrow lunch?" — you discover the gap. The model has no idea what is in your systems. For most of 2023 and 2024, the answer was: write a custom integration. Plug Claude into your POS via the POS's API. Plug it into your accounting via QuickBooks' SDK. Plug it into your reservation system via a third-party connector. Twenty integrations later, you have a brittle pile of glue and an engineer whose entire job is to keep it from falling over. ## What MCP actually is [Model Context Protocol](https://modelcontextprotocol.io) is an open standard, originally proposed by Anthropic in November 2024 and now adopted across the major frontier-model vendors. It defines a small, opinionated way for any tool — a database, a SaaS app, an internal script, a CSV — to expose itself to a language model. The model speaks one language (MCP). Each tool speaks one language (MCP). The integration cost stops compounding. > MCP is to AI agents what USB-C is to laptops. It is not glamorous. It is the reason you stopped travelling with seven different cables. Concretely, an MCP server exposes three things to a model: **resources** (read-only data the model can consult — your menu, your customer list), **tools** (functions the model can call — create a reservation, send a WhatsApp message, write to inventory), and **prompts** (reusable instructions that scope what the model should do with a given resource or tool). A model with MCP access reads the resources, picks the right tool, calls it with the right arguments, and reads the response — all while keeping the business's data inside the business's perimeter. ## Why it matters for an operating business The day-to-day reality of running a hospitality group, a clinic, or a retail operation in Muscat is that twelve tools hold pieces of the same truth. The same guest is a row in your POS, a row in your CRM, a thread in your concierge WhatsApp, and a note in someone's phone. Before MCP, building an AI concierge that understood all four of those things was a months-long custom build that broke whenever any of the four tools changed. After MCP, it is a configuration exercise. And the standard is getting more useful, not less. In May, Anthropic shipped [self-hosted sandboxes and MCP tunnels](https://claude.com/blog/claude-managed-agents-updates) for Claude Managed Agents — meaning a persistent agent can now reach back into a business's private network without exposing it to the public internet. That single update closes the biggest remaining objection to running AI agents over regulated data. ## The 4D framework, mapped to MCP Our four operating disciplines apply to MCP work directly: - **Delegation.** Decide which questions Claude should answer using which MCP server. Not every read is worth a tool call. - **Description.** The MCP *prompt* primitive is exactly this — telling the model how to use a tool, in writing, the way you would brief a thoughtful new colleague. - **Discernment.** Read the agent's tool calls the way you'd read a junior engineer's pull request. Eval before launch. Spot drift early. - **Diligence.** Audit the MCP server's surface area before granting access. The smallest tool set that works is the right one. ## What to do with this If you read one official MCP resource, make it the [MCP introduction](https://modelcontextprotocol.io/introduction). If you want a worked example, the [reference servers](https://github.com/modelcontextprotocol/servers) repo is the canonical place. And if you would rather skip the standard and have us build the right one for your operation, that is exactly what a Discovery Phase decides: [start one](https://www.orfloat.com/contact). — Akram Ahmed, CTO · Orfloat · Muscat ## References 1. Anthropic. *Introducing the Model Context Protocol.* 25 November 2024. [anthropic.com/news/model-context-protocol](https://www.anthropic.com/news/model-context-protocol) 2. Model Context Protocol. *Specification & documentation.* [modelcontextprotocol.io](https://modelcontextprotocol.io) 3. Anthropic. *New in Claude Managed Agents: self-hosted sandboxes and MCP tunnels.* 19 May 2026. [claude.com/blog/claude-managed-agents-updates](https://claude.com/blog/claude-managed-agents-updates) 4. Model Context Protocol. *Reference servers.* [github.com/modelcontextprotocol/servers](https://github.com/modelcontextprotocol/servers) [← Back to Notes](https://www.orfloat.com/notes) --- # Privacy Notice > Source: https://www.orfloat.com/privacy > Updated: 2026-05-25 # Privacy Notice Studio · Privacy Notice This notice describes how Orfloat collects, uses, and protects the personal data you provide through this website. It is designed in line with the principles of Oman's Personal Data Protection Law (Royal Decree 6/2022) and will be revised as our practices evolve. Last updated: 24 May 2026 · Controller: Afraa & Mufassir LLC, Muscat, Sultanate of Oman. ## 1. Who we are "Orfloat" is a trading name of Afraa & Mufassir LLC, a limited liability company registered in the Sultanate of Oman (Commercial Registration 1504141, License L3717644), with its registered address at PO Box 1, P.C. 411, Sur, Oman, and principal operating base in Muscat. We are the data controller for the personal data described in this notice. ## 2. What data we collect When you submit the contact form on [/contact](https://www.orfloat.com/contact), we collect your name, company, email address, and the message you write. We do not currently set analytics, marketing, or advertising cookies on this site. ## 3. Why we collect it We process the data you submit only to reply to your enquiry, schedule an introductory call where appropriate, and — if both sides see a fit — to enter into an Engagement Letter. We do not sell, rent, or share your data with third parties for their marketing purposes. ## 4. Where it is stored Enquiry data is stored in the founders' corporate email inbox and may be processed using standard business productivity tools (e.g. email, document storage). We use providers whose facilities may be located outside Oman; where this is the case, we rely on the operators' published security and data protection commitments. ## 5. How long we keep it We retain enquiry correspondence for as long as is reasonably necessary to respond, to operate any resulting engagement, and to comply with our legal and tax obligations. You can ask us to delete your enquiry at any time by writing to [orfloat.studio@gmail.com](mailto:orfloat.studio@gmail.com). ## 6. Your rights Subject to Oman's Personal Data Protection Law and any other applicable law, you have the right to ask us to confirm what personal data we hold about you, to correct it if it is inaccurate, to delete it where appropriate, and to withdraw any consent you have given. Please write to [orfloat.studio@gmail.com](mailto:orfloat.studio@gmail.com) and we will respond within a reasonable period. ## 7. Security We take reasonable technical and organisational measures to protect personal data against accidental or unlawful access, loss, or disclosure. No system is perfectly secure; please do not share information through this site that you would not be comfortable sending by email. ## 8. Changes to this notice We may update this notice from time to time. The "last updated" date at the top reflects the most recent revision. Material changes will be announced on this page. ## 9. Contact For any question about this notice or about how we handle your personal data, please write to [orfloat.studio@gmail.com](mailto:orfloat.studio@gmail.com). --- # Terms of Use > Source: https://www.orfloat.com/terms > Updated: 2026-05-25 # Terms of Use Studio · Terms of Use These terms govern your use of this website. They do not create any client engagement — engagements between Orfloat and a client are governed by a separately signed Engagement Letter. Last updated: 24 May 2026 · Operator: Afraa & Mufassir LLC, Muscat, Sultanate of Oman. ## 1. Operator This website is operated by Afraa & Mufassir LLC, a limited liability company registered in the Sultanate of Oman (Commercial Registration 1504141, License L3717644), trading as "Orfloat". ## 2. Purpose of this website The website provides information about Orfloat's services and a contact form through which you may enquire about an engagement. Information published here is provided for general information only and does not constitute professional, legal, or commercial advice. ## 3. No contract formed by browsing or by enquiry Browsing this website, downloading any material from it, or submitting the contact form does not create a contractual relationship between you and Orfloat. Any engagement is conditional on a signed Engagement Letter setting out the scope, deliverables, schedule, and fees of the work. ## 4. Intellectual property All content on this website, including text, images, the Orfloat wordmark, page structure, and the underlying source code, is the property of Afraa & Mufassir LLC or of its licensors, and is protected under applicable intellectual property laws. You may view and print pages of this website for personal, non-commercial reference. Any other use requires our written permission. Third-party names referenced on this site — including Anthropic, Claude, Claude Code, Model Context Protocol, and others — are the trademarks of their respective owners. See our [Trademarks & Attribution](https://www.orfloat.com/trademarks) notice for details. ## 5. No warranties This website is provided on an "as is" basis. We make no warranty, express or implied, that the information on it is accurate, complete, current, or fit for any particular purpose. We reserve the right to amend the site, including these terms, at any time. ## 6. Limitation of liability To the maximum extent permitted by law, Afraa & Mufassir LLC, its founders, employees, and contractors will not be liable for any loss or damage — whether direct, indirect, consequential, or otherwise — arising out of your use of, or inability to use, this website. ## 7. Third-party links Where this site links to a third-party website, we do not control that site and are not responsible for its content, terms, or practices. The presence of a link does not imply endorsement or affiliation. ## 8. Governing law These terms are governed by the laws of the Sultanate of Oman. The courts of Muscat have exclusive jurisdiction over any dispute arising out of or in connection with them, without prejudice to any mandatory consumer-protection rights you may have where you live. ## 9. Contact Questions about these terms can be sent to [orfloat.studio@gmail.com](mailto:orfloat.studio@gmail.com). --- # Trademarks & Attribution > Source: https://www.orfloat.com/trademarks > Updated: 2026-05-25 # Trademarks & attribution Studio · Trademarks & Attribution Orfloat's work is built on top of, and inspired by, products and writing from other organisations. This notice acknowledges their trademarks and the parts of their thinking that show up in our practice. Last updated: 24 May 2026. ## Independence Orfloat is operated by Afraa & Mufassir LLC and is an independent software studio based in Muscat, Sultanate of Oman. Orfloat is **not** affiliated with, endorsed by, sponsored by, or partnered with Anthropic or any other third party named on this website, unless explicitly stated on a specific page. ## Anthropic trademarks "Anthropic", "Claude", "Claude Code", "Claude Constitution", "Model Context Protocol", and the Anthropic product names are trademarks of Anthropic, PBC. We reference these names nominatively — that is, to accurately describe the technology our work is built on — and not as a claim of affiliation or endorsement. ## The 4D framework The four disciplines we describe on [the homepage](https://www.orfloat.com/) — Delegation, Description, Discernment, and Diligence — are inspired by Anthropic's published material on Applied AI and prompting practice. The framing is theirs; the operational application inside our engagements is ours. Where we use the label "4D framework" on this website, it should be read as a credit to Anthropic's thinking and not as a framework we invented. ## Other third-party marks Names of other companies, products, and services referenced on this site are the trademarks or registered trademarks of their respective owners. Their inclusion is descriptive and does not imply any endorsement of Orfloat or vice versa. ## Orfloat marks "Orfloat" and the Orfloat wordmark are marks of Afraa & Mufassir LLC. Please do not reproduce them outside of factual reference to Orfloat without our written permission. ## Corrections If you are the owner of a trademark you believe is misused on this site, please write to [orfloat.studio@gmail.com](mailto:orfloat.studio@gmail.com) and we will respond promptly.