AI Engineer · Solo Builder · Backend Systems
I build AI-native systems, RAG architectures, and automation tools using modern backend + LLM stacks. From 0 to production in days.
Selected for TinyFish AI Accelerator. Built and shipped OmniProcure from concept to real users during a high-pressure sprint week.
Built OmniProcure end-to-end during the sprint. API integration, AI layer, auth, and deployment all shipped under time pressure.
Integrated OEM Secrets API to surface real-time supplier data across 140+ hardware distributors including Mouser, DigiKey, LCSC.
Natural language queries like "RP2040" resolve to exact SKUs like SC0914 and match across suppliers automatically.
Shipped to 10+ hardware manufacturers (0 to 35 years experience). Collected live feedback and iterated in real time.
Real systems, real deployments, real users. Not tutorials.
AI-powered procurement automation that streamlines hardware sourcing. Uses AI for intelligent part number resolution (e.g. "RP2040" to SC0914), integrates 140+ distributors via OEM Secrets API, and serves per-supplier cards with quantity selectors. Guest flow with OAuth gating for PO generation and cron-based async monitoring for automated workflow execution.
High-performance RAG-based semantic search engine optimized for CPU inference. BGE-M3 embeddings into ChromaDB, returning ranked context-aware results via FastAPI at ~1s latency on CPU.
Local-first persistent memory for AI agents enabling long-term contextual recall. LanceDB and Ollama embeddings for semantic retrieval across sessions. Built as a node for the Aden Hive agent framework.
Full-stack AI systems from embedding pipelines to deployed products.
Open to collaborations, contracts, and conversations.