<aside> ✨ a monthly newsletter about ML topics and papers. goes directly in your email inbox. subscribe by emailing anything to [email protected]

</aside>

👋 who am i? i’m arushi. i work on creating the next generation of how we talk to machines and how they talk to us.

📧 want to sign up for monthly emails? drop me an email at [email protected]

🚪 find me on twitter, more of my stuff at amks.me

Latest Explainers

Monte Carlo Tree Search

ML at Scale: Pipeline Parallelism

ML at Scale: Tensor Parallelism

ML at Scale: Data Parallelism

How do Mixture of Expert Models Work?

Notes: Steve Brunton’s Introduction to RL

A Taxonomy of Reinforcement Learning

Experimental Work

Research Logs

Some selected works:

• Implementing LLM MCTS-based Inference for Code Generation
• Measuring the Arithmetic Capabilities of Frontier Models
• Measuring Few-Shot Variance on Frontier Model Canonical Text Evaluations