Shipped products and autonomous pipelines — built lean, deployed at AI-speed. Hover a tile, then dive into the architecture.
Typed AI decisions for staffing — the model recommends, a human decides
Ranks people for open tasks with typed decisions that return a probability for every option. Code applies the hard rules, a ModernBERT student distilled from an open-weight teacher answers five bounded questions, and a manager makes every final call.
Local-First AI Financial Agent
A privacy-first personal finance agent that keeps 100% of your financial data on-device in SQLite — no cloud backend. An OpenAI function-calling agent tracks debts and pushes proactive briefings, with every database mutation gated behind a Review & Confirm boundary.
Ultra-lightweight streaming cockpit
A Rust-powered desktop cockpit for Twitch & YouTube streamers — live system telemetry, multi-platform chat velocity, and automated OBS scene sync, all inside a ~152 MB / 1.8% CPU footprint.
Autonomous content scheduler
A Python autonomous agent running on GitHub Actions that schedules and publishes technical content on a cron — drafting every post through the Gemini API. Zero servers, zero manual posting.
Git-triggered article factory
A native Next.js pipeline that autonomously writes, formats, and deploys Markdown articles straight to this site — Gemini drafts the MDX, GitHub Actions commits it, and GitHub Pages ships it.
I'm Yaseen Khatib — a Senior Full-Stack AI Engineer (MERN + TypeScript) who ships production AI systems solo: RAG pipelines, agent orchestration, real-time and edge architectures. Open to senior / lead roles and contract work, remote or on-site.