work stuff
Research

Siddhartha Rao Kamalakara

Twitter

Instagram

Scaling @ Runway

LLMs at Cohere Systems x Foundation models

Drug design at ProteinQure

Research @ FOR.ai

Learning Sparse Networks Using Targeted Dropout

I enjoy doing research but I don’t publish anymore. Here are a few things I worked on in the past

Scalable Training of Language Models using JAX pjit and TPUv…
Exploring Low Rank Training of Deep Neural Networks

https://joss.theoj.org/papers/10.21105/joss.01524

RL: Generic reinforcement learning codebase in TensorFlow

Outside of work, I enjoy playing new instruments poorly and singing, photography and filmmaking

Pictured - The Best City (Toronto)

A playlist close to me

born and raised