french_exec

5 ms·

1. 1.▲
   by french_exec 5mo ago

   "Unfortunately, GPT‑5.5 started training before we found the root cause of the goblins. When we began testing GPT‑5.5 in Codex, OpenAI employees immediately noticed the strange affinity for goblins, and we added a developer-prompt inst
2. 2.▲
   What a Null Result Taught Us About AI Agent Evaluation(clouatre.ca)

   1 points by french_exec 7mo ago|0 comments
3. 3.▲
   by french_exec 8mo ago

   Fair point. Probabilistic systems being unpredictable isn't news. The article's just noting current observability tools weren't designed for these failure modes.
4. 4.▲
   Why Your AI Agent Failed in Production(clouatre.ca)

   1 points by french_exec 8mo ago|2 comments
5. 5.▲
   Orchestrating AI Agents: A Subagent Architecture for Code(clouatre.ca)

   1 points by french_exec 8mo ago|0 comments
6. 6.▲
   RAG for Legacy Systems: 7,432 Pages to 3s Answers(clouatre.ca)

   4 points by french_exec 8mo ago|0 comments