May 9 · 3 min read · The day a billing agent refunded $42,000 by accident In February 2026, one of our internal agents — a refund-triage bot running on Claude Opus 4.7 — issued a perfectly polite, perfectly catastrophic chain of refunds totaling around $42,000 before our...
MMax commented
Apr 28 · 6 min read · The Silent Killer of Multi-Agent Systems Isn't the Model. It's Topology Mismatch. In the last 14 days, three things happened in AI agents that should have settled the reliability conversation. Instead, they revealed how badly we're framing it. Stanfo...
Join discussionApr 26 · 5 min read · Built in the open. Trained in public. Shipped with the spirit of the open-source AI community. A February Evening That Changed My Thinking On 17 February, I walked into an Anthropic event carrying o
Join discussion
Apr 9 · 5 min read · We Rebuilt Our RAG Pipeline 4 Times — Here's the Architecture That Finally Served 50K Daily Queries Under 800ms Our first RAG system hit 91% user satisfaction in demos and 34% in production. This is the brutal post-mortem of 4 rebuilds, 3 fired vendo...
Join discussion
Apr 1 · 5 min read · Every AI coding tool you use needs access to your code to function. Copilot reads your files for completions. Cursor indexes your project for context. LangChain traces log your prompts and outputs for
Join discussionMar 30 · 5 min read · In Part 4, we closed the loop on authentication and established our "Stop & Ask" mandate. We moved from a blank slate to a persistent session model. But a login screen is just a door—this session was
Join discussion
Mar 29 · 6 min read · Your LLM Is Lying to You Silently: 4 Statistical Signals That Catch Drift Before Users Do No 500 errors. No latency spikes. Just 91% of production LLMs quietly degrading — and your dashboards showing green the whole time. Here's the core tension I ke...
Join discussion
