1+ week, 2+ day ago   (803+ words) The schedule is a teaching box, not a production SLA, and it assumes one laptop plus one HTTP scoring endpoint. If the endpoint is slow, freeze Exercise 2 after five scored chunks and continue with the logged samples. Do not expand…...

2+ week, 1+ day ago   (982+ words) An agent-patch gate that never rejects a patch is a hypothesis, not a control. You cannot measure what it catches until you give it something to catch. So I stopped trusting real patches and started manufacturing failures. The method that…...

2+ week, 4+ day ago   (1390+ words) The Audience I Didn't Know I Had — Part One There is a lot of conversation right now about one developer coordinating forty coding agents. I am not going to pretend I know whether that is good or bad. It is…...

3+ week, 7+ hour ago   (965+ words) Building an AI feature is easy. Building a reliable multi-agent pipeline that coordinates four specialized AI agents, persists intermediate state, skips already-completed work on retries, and keeps the API responsive while the models think — that is the hard part. This…...

3+ week, 1+ day ago   (180+ words) Run this on any repo doing spec-driven development: Same rule, three specs, three wordings — and two meanings. 30 days from purchase is not 30 days from delivery. Nobody decided that fork. A coding agent paraphrased it into existence, and no human caught…...

3+ week, 2+ day ago   (910+ words) gex.live has an MCP server. Add it to Claude, Cursor, ChatGPT or any MCP client and the assistant can read the dealer-positioning archive and drive the backtesting Lab on your behalf. The one-line version is the title. Here is…...

1+ mon, 6+ day ago   (569+ words) The last lesson was about validating what a model hands you. The story behind it: a set of prompts that had returned real, criteria-matched vendors for weeks came back in staging with placeholder junk, literally the words Vendor A, Vendor…...

1+ mon, 6+ day ago   (357+ words) Most AI agent tutorials show you the same pattern. You write a while True: loop. The AI calls a tool, gets a result, and calls another tool, again and again, until it decides to stop. It looks simple. It looks…...

1+ mon, 2+ week ago   (904+ words) The temperature in your database was, at some point, a current in a wire. Not as a metaphor. Actual milliamps through copper. Almost everything your code knows about a physical process traveled as an electrical current through a cable before…...

2+ mon, 5+ day ago   (595+ words) un Trivy against almost any vendor container image and you'll get a wall of findings. Most of them don't matter, the vulnerable code path is never executed in your deployment. The problem is that knowing that has traditionally lived in…...