Writing, reviewing & teaching.
Technical reviewing for Manning Publications, a body of long-form articles written for clients, an ongoing public series on production AI, and the occasional classroom — all on the same questions about what makes AI systems actually work.
Evaluating AI Systems
Technical review of Panos Alexopoulos's book on evaluating AI systems — retrieval quality, generation, and end-to-end measurement.
Rearchitecting LLMs
Technical review of Pere Martra's book on the architectures behind modern LLMs and how they evolve in production.
Walma.ai — thought leadership library
Wrote Walma's full library of long-form thought leadership — establishing the company's technical authority and the foundation for downstream content.
- EU data sovereignty and the legal architecture of European AI
- RAG pipeline architecture — what production retrieval actually requires
- Shadow AI risk and how enterprises lose control of model usage
- The EU AI Act — practical implications for builders, not lawyers
- What production-grade enterprise AI looks like (and what it doesn't)
AI system design, evaluation, and transparency
Ongoing series on reliability patterns, evaluation strategy, retrieval, and the gap between AI demos and production systems.
- 170K+ impressions, 112K+ reach across the broader series
- Featured viral pieces at 36K and 124K impressions
- Structural patterns from the viral posts now applied to client content systems
Teaching LLMs, MCPs, and AI fundamentals
Taught core AI concepts to 30+ classmates at IHM Business School — model fundamentals, retrieval, agentic patterns, and the Model Context Protocol.