inhenn

[inhenn @at_rubick]

https://t.co/nnIIfQVpZ2

[inhenn @at_rubick]

The training run finished. The loss looks great. But the model is broken. The worst AI training bugs don’t crash. They hide in parts of the training pipeline, waste days of human debugging and GPU time.

Check out TrainingDxBench for this benchmark. https://t.co/VmnpiXeo1l

[inhenn @at_rubick]

If you want to automate a computer use task, just record yourself doing it once—the agent can then follow your trajectory to complete the same task or adapt it to similar tasks. Excited to share our work: https://t.co/QsXLGIXohd

[inhenn @at_rubick]

@swyx

[Learn Prompting @learnprompting]

🚨Announcing The Prompt Report🚨

A 76-page survey of 1,500+ prompting papers, analyzing EVERY prompting technique, Agents, & GenAI

Led by @learnprompting, and folks from @OpenAI, @Microsoft, & @UofMaryland

Here’s what we found & the 58 prompting techniques you should know👇🧵

[inhenn @at_rubick]

[WEBINAR] Introduction to #MachineLearning for Quantitative Finance
Register: https://t.co/GF1m8vx4mi 来自 @quantinsti