AI

I build practical AI systems. Not demos, not wrappers around ChatGPT. My focus is on custom RAG pipelines, agent orchestration platforms, and context-aware integrations that solve real problems.

Custom RAG Systems

I design and build Retrieval-Augmented Generation pipelines that ground LLM responses in your actual data. This means vector embeddings, intelligent chunking strategies, and retrieval layers that give AI models the context they need to produce accurate, domain-specific answers instead of hallucinations.

AI Agent Orchestration

Single-prompt AI is table stakes. I build multi-agent orchestration platforms where specialized AI agents collaborate on complex tasks: coordinating workflows, delegating subtasks, and synthesizing results. This is where AI moves from a tool you talk to into a system that works for you.

Context Window Engineering

I've built systems that dynamically inject real-time data into an AI's context window at the moment a conversation begins. For example, my web-based voice assistant built with Vertex that pulls my latest listening data from the Spotify API and recent reads from the Hardcover App API, weaving them into the AI's awareness before you even say a word.

AI Observability

I also built JTPCK, an AI observability platform that gives your AI usage data back to you as an API. If you can't measure it, you can't improve it.

If you'd like to talk about AI work in Boston, send me an email.