AI Builder | Product Engineer | Agent Systems Architect
Building intelligent systems that ship. From multi-agent orchestration to full-stack SaaS � I design, build, and deploy AI-native products end to end.
I work at the intersection of AI engineering and product. My focus is on systems that are useful in production � not demos, not proofs of concept. Agents that run autonomously, pipelines that scale, tools that people actually use.
Current obsession: Multi-agent architectures � designing fleets of specialized AI agents that coordinate, self-monitor, and produce real output without human babysitting.
AI/ML Python · PyTorch · LangChain · RAG · Vector DBs · Fine-tuning · Embeddings
Backend FastAPI · Node.js · Express · PostgreSQL · Redis · WebSockets
Frontend Next.js · React · TypeScript · Tailwind
Infrastructure Docker · GitHub Actions · Vercel · Linux · Shell scripting
Agent Systems Multi-agent orchestration · Cron scheduling · Memory systems · Tool use
| Project | What It Does | Stack |
|---|---|---|
| rag-pipeline | Production RAG system with chunking strategies, vector store abstraction, and retrieval evaluation | Python · FAISS · FastAPI |
| agent-framework | Lightweight agent orchestration library with tool registration, memory, and planning loops | Python |
| ai-toolkit | CLI for everyday AI/ML tasks � embeddings comparison, token counting, model output diffing | Python · Click |
| startup-dashboard | SaaS metrics dashboard tracking MRR, churn, LTV, CAC with real-time charts | Next.js · TypeScript · Recharts |
| prompt-eval | Prompt engineering toolkit � version control, A/B testing, and performance tracking for prompts | Python · FastAPI |
I don't believe in building things that only work in a notebook. Every project I ship has a CLI or an API, handles errors gracefully, includes tests, and comes with documentation that assumes the reader's time is valuable. I care about code that runs at 2 AM without waking anyone up.
Open to collaborations on AI tooling, agent systems, and developer infrastructure.