Parv Bhargava·Aug 14Your LLM Judge Scores 4.2 Out of 5 and Catches NothingA walkthrough of how production evaluation actually fails — with a worked example, the statistics that matter, and the failure modes that…
Amit.Kumar·May 27Testing Non-Deterministic Systems: How to Test Your AI Agent When the Output is Never the SameYour unit tests pass. Your integration tests pass. And then your agent confidently tells a user the wrong thing. Here’s how to actually…
Sundar Ram Subramanian·May 23Multi Agentic Merchandiser | Caught in Flight, Found in TracesIn the last article, we saw the Multi Agentic building blocks of the Multi Agentic Merchandiser. Here, I am sharing the specific failure…
Sundar Ram Subramanian·May 21Agentic AI | Planning & Multi-Agent WorkflowsThis article is about the remaining two patterns: Planning and Multi-agent collaboration.
Saranya Sankar·Mar 30Evals are a PM’s best tool for shaping agent design.If you’re a PM working on AI products and have the opportunity to get involved in evals then take it.
