Head of AI & Engineering @ Workhelix · Founding Engineer · O'Reilly Author
I am Head of AI and Engineering and the founding engineer at Workhelix, where I build enterprise agentic and LLM platforms from zero. My work spans LLM serving and inference on self-hosted GPUs, agentic workflows, MCP integrations, RAG, evaluations, and the developer tooling around them. That platform now runs inside Fortune 50 companies like Nasdaq and John Deere.
I am also the O'Reilly author on AI agents and agent memory, an AEA-published researcher, and a former university machine learning instructor. I like turning hard systems into clear ideas.
- 🔭 Building agent platforms at Workhelix and writing O'Reilly's book on Agent Memory
- 💬 Ask me about agentic systems, LLM inference, MCP, RAG, and taking GenAI from prototype to production
- 📫 [email protected]
- mltrack — lightweight experiment tracking for machine learning
- langgraph-redis-workshop — building stateful AI agents with LangGraph and Redis (ODSC West 2025)
- what_are_ai_agents — companion code for my O'Reilly report on AI agents
- opt_and_sim_with_llms — LLM optimization and simulation, guest lecture at the Wharton School
- Agent Memory — O'Reilly book, in progress
- Managing Memory for AI Agents — O'Reilly report (2025)
- What Are AI Agents? — O'Reilly report (2023)
- Extending "GPTs Are GPTs" to Firms — AEA Papers and Proceedings (2025)
More at econoben.dev/publications
- The Evolution of AI Agents — Into the Hopper Podcast (2026)
- Building Stateful AI Agents with LangGraph and Redis — ODSC West (2025)
- AI Needs Memory: Here's How It Works — MLOps Community (2025)
- Building With AI: Quick POCs with LLMs — Wharton Guest Lecture (2024)
- A Normie Approach to Validating LLM Outputs — AI.Science (2023)
- Building an HTTPS Model API for Cheap — NormConf (2022)
More at econoben.dev/talks
Turning hard systems into clear ideas.