Shreyans Jain
Interpretability Researcher at Self
Bengaluru, Karnataka, India
Summary
🤩
Rockstar🎓
Top SchoolShreyans Jain is an interpretability researcher and experienced data scientist with 8+ years building production ML systems—OCR, search, recommendation, and time-series forecasting—out of Bengaluru. He publishes and maintains open-source tooling for visualizing neural network training dynamics (ICLR 2025 blogpost and a public GitHub repo) and contributes to libraries like Minari and AgileRL. His recent independent research includes arXiv work and accessible explainability write-ups that bridge practical model diagnostics with theoretical insights. Previously he delivered analytics and ML solutions across enterprises like GEP, BookMyShow, Hotstar and Fractal, and has applied psychometric frameworks to evaluate LLM behavior in applied settings. A civil-engineering graduate by training, he pairs strong applied engineering chops with a knack for clear visual and experimental communication of model behavior.
8 years of coding experience
6 years of employment as a software developer
Bachelor of Technology (BTech), Civil Engineering, Bachelor of Technology (BTech), Civil Engineering at Malaviya National Institute of Technology JaipurHigh School, Mathematics and Statistics, High School, Mathematics and Statistics at Maheshwari Public School
English
Github Skills (19)
gymnasium10
evolutionary-algorithms10
reinforcement-learning10
pettingzoo10
q-learning10
multi-agent-reinforcement-learning9
face9
datasets9
machine-learning9
multi-agent9
deep-learning9
hyperparameter-tuning9
deep-reinforcement-learning9
python8
pytorch7
Programming languages (4)
Github contributions (5)
shreyansjainn/Minari
May 2023 - Nov 2023
A standard format for offline reinforcement learning datasets, with popular reference datasets and related utilities
Contributions:38 pushes, 8 branches in 5 months
shreyansjainn/visualizing-training
Oct 2024 - Dec 2024
Visualizing Training Dyanmics of a Neural Network
Contributions:3 reviews, 11 PRs, 35 pushes in 1 month
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.