End-to-End RL Post-Training

4 pointssarthakaggarwal1y ago2 comments

https://maxreward.vercel.app/

Reinforcement Learning is fun, but building the RL pipeline is not fun. We bring the fun back in RL.

2 comments

2 comments · 2 top-level

Pretty good for the devs I feel. Takes up a lot of my inertia to go away from prompt engineering to something that actually works!

Amazing work!

j / k navigate · click thread line to collapse