The human behind the systems.
One engineer, five shipped products, ninety-plus articles — and a conviction that vision, not syntax, is the scarce skill.
Who is Yaseen Khatib?
Yaseen Nurmahammad Khatib is a Senior Full-Stack AI Engineer based in Hyderabad, India, who builds and ships autonomous AI products end-to-end — Agentic RAG pipelines, LLM orchestration, and the high-throughput MERN systems they run on. In the last twelve months he designed, built, and shipped five production products entirely solo.
His thesis, argued across the Founder's Log series, is that AI has made implementation abundant and judgment scarce: the engineer who can hold an entire system in their head — data contracts, trust boundaries, unit economics — now ships the volume of a full squad. Every product on this site was built under that operating model, with Claude as the workforce, MCP as the wiring, and every architectural decision made by one human.
- name:
- Yaseen Nurmahammad Khatib
- role:
- Senior Full-Stack AI Engineer
- base:
- Hyderabad, India (IST) · remote, hybrid, or on-site
- products_shipped:
- 5 solo — streamerOS, IntegrateX, Sable, + 2 pipelines
- articles_published:
- 90+ on AI systems architecture
- operating_model:
- one architect directing Claude + MCP
- availability:
- open to Lead / Senior Full-Stack & AI roles
- email:
- contact@streamerosai.com
What has he built?
- streamerOS — a Rust + Tauri desktop cockpit for streamers — on-device chat ingestion, telemetry, and OBS automation at a fraction of Electron's memory.
- IntegrateX — a React Flow workflow engine whose schema-aware serialization adapter cut payloads by 94%.
- Sable — a local-first AI finance agent — all data on-device, and the model proposes while only a human commits.
- The Zero-Cost AI Blog Writer — an autonomous pipeline that drafts, commits, and deploys articles (with covers) daily for $0 of infrastructure.
- The LinkedIn Pipeline — an outreach machine that turns shipped work into scheduled LinkedIn content and tracks its state in Git.
Every claim links to a full architecture teardown in /products, and the thinking behind the code fills the blog.
How does he work?
Architecture first: the data contract, the trust boundaries, and the cost model are fixed before the first commit — then Claude Code, wired to real systems over MCP, fills the layers at a speed no team matches. The boundaries are non-negotiable: in Sable the model can propose but never touch the money; in streamerOS chat never leaves the user's machine; in the content pipelines a human curates every queue. Leverage without authority. The full playbook is documented lesson by lesson in the Claude Code roadmap.
Is he available?
Yes — open to Lead and Senior Full-Stack / AI Engineering roles, remote, hybrid, or on-site. Based in Hyderabad (IST), comfortable in an office, and effective across global time zones; the pipelines on this site work while he sleeps, which is rather the point. Engineering leadership evaluating him should start with the interview brief.