This is an actual f*cking goldmine
Anthropic just shared one of the best tricks for Claude Code. The idea: run Sonnet, Haiku 5.5 and Opus 5.5 at once, each one doing only the job it's best at. Better results and way fewer tokens, because every task lands on the right model.
Anthropic engineer, Addy Osmani (ex-Google)
"Google hired me at $500K–$1.2M a year because I know how to make AI agents write code that doesn’t break." In this 41-minute workshop he shows exactly how to build that system from scratch. Prompts → Agent Skills → Harness →
Karpathy's agent ran ~700 experiments in 2 days.
The trick costs $0 no new model. no team. one rule: every change gets a 5-minute test, and anything that doesn't beat the score gets reverted so i turned that rule into a graph for normal work: → opus writes the check first. one
This is the most DANGEROUS thing on the whole internet right now
The tools Goldman Sachs, Point72 & Man Group built for their own desks, plugged into Opus 5.5 = a 24/7 QUANT DESK that runs on your laptop i plugged them into Claude Code & built a COMPLETE AI trading setup from
You have reached the end of the archive
All of Claude Opus 5