IymanAhmed
From idea to intelligence.
Code. Evaluate. Ship.
I ship production AI systems: agent orchestration, RAG with citations, ranking engines, legal analyzers, and financial intelligence tools.
Stockton, CA — SF Bay Area
01 / Selected work
Built to solve.
Engineered to ship.
From parallel agents to evidence-backed AI. Explore the systems, the architecture, and the results.
Profile
What I bring.
What I build
- Multi-agent systems that research, score, and synthesize in parallel
- Compliance copilots that separate model extraction from deterministic calculation
- RAG pipelines with citations, hybrid search, and reranking
- Ranking & evaluation engines with explainable, per-dimension scores
- MLOps: eval harnesses and CI gates that block quality regressions
- Full products — FastAPI backends, streaming UIs, live deployments
Domains shipped
How I work
Eval-driven
Every system ships with metrics; CI gates block silent regressions.
Grounded, not guessing
Citations, guardrails, and abstention over confident hallucination.
Production-first
Deployed, latency-aware, end-to-end — ingestion to interface.
Stack
Python · TypeScript · LangChain · LangGraph · Anthropic · Groq · OpenAI · FastAPI · Postgres/pgvector · SQLite · ChromaDB · RAGAS · MLflow · Prometheus · Docker · Terraform · GitHub Actions · Vercel · Fly.io · Next.js · SHAP · PyTorch
Contact
Start a build.
Send a note and I’ll reply with next steps.
Response time
Usually within 24 hours.