file_000000006a1451f7a182aa4da6db7e10_conversation_id=67e83ff7-57f4-800e-8450-bfdb49d1355c&message_id=6592540b-fc95-4858-b303-ffdbaa9cadab.PNG

Hey there 👋 I'm Gaurang

I'm a Research Engineer at RTX Corporation, where I work on agentic systems, reinforcement learning, and LLM infrastructure. I have a particular interest in scaling agentic systems, models playing game (long horizon planning) and inner working of LLM (mechanistic interpretability)

prompt.utils

Latent Space

https://dynalist.io/d/2uxXpkcL6M-2JR_JuUpWFyNx

Blogs

:linkedin_icon-svg: LinkedIn | :1707226109new-twitter-logo-png: Twitter | :githublogo: GitHub

Why this website?

I created this website to document my research, learning, and things I find interesting—from RL and agentic systems to satellite tech and control theory. Most content lives in [public_cache], updated frequently as I learn.

| DL blogs #DL → BackPropogation: Math & Intution → Do DL models overfit?

→ Neural Collapse & Long-Tail in NN
System Design
→ ‣
LLMs
→ ‣
University + SpaceTech #misc
→ Controls: AODS & Thrust Vectoring for sounding rockets
→ SAR: Interpretability & Fusion

Previous Experience

Achievements

  1. Filed 2 patents for work at Raytheon and ISRO
  2. Co-founded Space Research Group STeRG at university (MIT Pune)
  3. Launched payload on ISRO's POEM-4 mission (Dec 2024)

Get in Touch

📩 Email | :linkedin_icon-svg: LinkedIn | :1707226109new-twitter-logo-png: Twitter | :githublogo: GitHub