tag
11 items · 7 projects · 4 feed items
Co-founded an end-to-end GenAI platform that automates the innovation workflow for Consumer Packaged Goods companies, from idea prototype through to production-ready concept. Full-stack React + Python + LLM integration, shipped to early customers.
Designed an AI-assisted review workflow that checks proposed event schemas against documented contracts and produces a structured, evidence-backed review.
Designed Model Context Protocol integrations that give AI agents structured access to location, routing, and traffic-analysis capabilities.
Contributed to an agent toolkit that lets LLM applications interact with mapping capabilities through structured, typed tools.
Designed a multi-agent prototype for answering traffic-analysis questions with scoped tools, structured evidence, and a single grounded response.
Hackathon prototype that turns an address into a structured neighbourhood profile using a multi-step, tool-using AI workflow.
Designed a planner-executor system that converts natural-language location requests into grounded, multi-step tool calls and explainable results.
How to make a 2-minute walkthrough of a search system with React, Remotion, and Claude Code—plus the small set of habits that made the process work.
Most agent evals measure the wrong thing. Output quality alone misses 80% of production failure modes. The five eval categories I now run on every agentic system, the LLM-as-judge gotchas, and how to ship a useful eval harness in a week.
Tool naming, description writing, argument schemas, error envelopes, idempotency. Lessons from authoring two production MCP servers, with the patterns I now reach for and the mistakes I've stopped making.
Counter-narrative to 'agents everywhere'. Multi-agent systems pay a real latency, complexity, and observability cost. Here's the decision framework I use to decide whether to split, and three concrete cases where a single call beats N agents in a graph.