SDET·Oct 25Improving AI Agent Reasoning with In-loop PlanningIn-loop planning is a technique that allows an AI agent to plan, act, observe the results, and revise its plan repeatedly while solving a…
Agent Native·Sep 13Your Agent Is Stuck in Fractal Reasoning BasinIf you are building an AI agent, your eval stack probably measures success rate, tokens, latency, cost, and maybe tool-call accuracy.A response icon1
Micheal Lanham·Sep 2Your Agent’s Reasoning Lives in Your Code, Not in Your PromptTips 65 to 70 of 100. A forward pass evaluates. It does not prune, backtrack or iterate, because none of those are things a single pass can…
Nishu singh·May 18Why most “AI analytics” agents hallucinate — and the two bets that fixed it for usA field report from building a multi-agent reasoning system on top of Databricks, Unity Catalog metric views, and LangChain’s deepagents…
