Knowledge RAG
A full-stack RAG application with multi-kind ingestion, unified retrieval, and per-user chat history — all running locally with Ollama and ChromaDB.
Works
A showcase of AI systems, intelligent applications, and tools built to explore ideas, solve real-world challenges, and push the boundaries of what’s possible with AI.
A full-stack RAG application with multi-kind ingestion, unified retrieval, and per-user chat history — all running locally with Ollama and ChromaDB.
Grounded retrieval with citations and confidence scoring, built for teams who can't ship hallucinated answers.
Planner → researcher → writer agents that produce cited briefs from a single prompt.
Turns PDFs, scans, and slides into structured, queryable data with section awareness.
Repo-aware context engine that lets a model reason about unfamiliar code without copying it into a vendor cloud.
Long-running automations that combine deterministic steps with agentic decision points.
Regression gates, rubric scoring, and human-in-the-loop review for LLM features.
Model routing, quantization, and offline inference for serious local deployments.
Email triage, calendar shaping, and daily briefs stitched into one quiet surface.
Let's build something.
If you're working on something that needs real AI engineering — not just a demo — let's talk.