hello!

hi, my name is Himanshu! i am a machine learning engineer stuck in the constant loop of learning new things and building cool stuff!

currently, i am a lead machine learning engineer at PrudentBit! previously, i was a researcher at Lossfunk working on reducing compounding rollout error and its induced distributional shift and studying RL generalizability on benchmarks such as Atari-100K.

i love deep learning, LLMs and am always looking to build cool projects on these, so ping me up on any of the below if you would like to collaborate on something!

• [@retr0sushi_] : twitter
• [@himanshuhasnoenemies.] : discord
• mail

i am currently working on the following topics :

• GPU programming
• reinforcement learning
• world models
• some interpretability and ai safety stuff as well
• also teaching ML and math!
• sometimes i run cool experiments on my own time and put them up on my github here : github

talks :

• Transformers to Mamba at AI4Bharat by IIT Madras watch here!

blogs section (i don't write much but most of what i become fascinated by, i share on my twitter) :

• residual streams in transformer models
• the platonic representation hypothesis
• my mech interp beginner questions

thanks for visiting!