Satyam Goyal

Fremont, CA
Hi! I’m Satyam Goyal, an incoming Masters student at the University of Michigan, Ann Arbor, studying Computer Science. I’m excited about GenAI applications specifically making agents that can rapidly learn. I’m particularly interested in RL and self-improvement in LLMs. I’ve worked on exciting projects like developing an AI-based academic assistant, improving colorblind image correction algorithms, and creating benchmarks for linguistic reasoning in LLMs.
When I’m not immersed in tech, I love playing the Hindustani flute and enjoy a good game of table tennis. Thanks for getting to know a bit about me!
news
Jun 15, 2025 | Attended YC Startup School and got to hear from Sam Altman, Elon Musk, Satya Nadella, and Aravind Srinivas. |
---|---|
May 19, 2025 | Began my position at IBM in NYC. |
May 15, 2025 | Working on startup called Hermes to make an internal agent builder that uses natural language. |
May 01, 2025 | Completed Bachelors of Science in Computer Science at the University of Michigan, Ann Arbor. |
Apr 20, 2025 | Accepted to AI Startup School from Y Combinator, will be in the bay area in June. |
Apr 10, 2025 | Received honors distinction at University of Michigan Undergrad. |
Feb 25, 2025 | Accepted to University of Michigan for 1 year Masters in Computer Science. |
Jan 08, 2025 | Submitted preprint of IOLBENCH, a benchmark to test LLM performance on linguistic reasoning. arXiv:2501.04249 |
Dec 10, 2024 | Accepted a position for IBM Research as a Generative AI and Cloud Intern. Will be working on improving model serving on their platform called llm-d. |
Aug 13, 2024 | Updates to Tara - Improved inference speeds and better student Audit Parsing |
Jul 13, 2024 | Working with Dr.Dan at IBM on Benckmarking LLMs on Linguistic Problems Sets |
May 13, 2024 | Started Intership Position with Optiwise.ai! Working on automatiating marketing strategies with LLMs. |
selected publications
- IOLBENCH: A Benchmark to Test LLM Performance on Linguistic ReasoningarXiv preprint arXiv:2501.04249, 2025
- Yoga pose perfection using deep learning: An algorithm to estimate the error in yogic posesJournal of student Research, 2021
- EnColor: Improving Visual Accessibility with a Deep Encoder-Decoder Image Corrector for Color Vision Deficient IndividualsIn Proceedings of the AAAI Conference on Artificial Intelligence, 2024