Yash Mehta
Research Intern at Microsoft
Baltimore, Maryland, United States
Summary
👤
Senior🎓
Top SchoolYash Mehta is a PhD student in Computational Cognitive Science at Johns Hopkins University with a decade of experience building bridges between neuroscience and machine learning. His work spans academic labs (HHMI Janelia, UCL, Freiburg) and industry (Amazon, Sakana AI, Microsoft), focusing on representations in biological and artificial neural networks and autonomous scientific discovery. He co-authored ICLR work on efficient AutoML / NAS, contributed DrNAS implementation to the widely used NASLib, and applies probabilistic and evolutionary methods to LLM-driven idea generation. Equally comfortable writing production code and designing biologically grounded models, Yash blends rigorous empirical research with hands-on engineering to push scalable, interpretable AI systems.
10 years of coding experience
4 years of employment as a software developer
BITS Pilani, Birla Institute of Technology and ScienceJohns Hopkins UniversityHigh School Diploma, Science, High School Diploma, Science at National Public SchoolResearch Thesis , Natural Language Processing, Deep Learning, Research Thesis , Natural Language Processing, Deep Learning at Nanyang Technological University Singapore
Japanese, Hindi, English
Stackoverflow
1reputation
0reached
0answers
0questions
Github Skills (10)
algorithm10
algorithms10
pytorch10
machine-learning10
deep-learning10
python10
nas10
neural-architecture-search10
implement10
git8
Programming languages (3)
Github contributions (5)
automl/NASLib
Nov 2020 - Mar 2022
NASLib is a Neural Architecture Search (NAS) library for facilitating NAS research for the community by providing interfaces to several state-of-the-art NAS search spaces and optimizers.
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:1 review, 145 commits, 38 PRs in 1 year 4 months
Contributions summary:Yash primarily contributed to the development and implementation of the DrNAS optimizer, which is part of the neural architecture search (NAS) library. Their work involved modifying the DARTS optimizer, incorporating Dirichlet distributions (DrNAS), and integrating a KL divergence regularization term. They also added functions for discretizing the architecture, demonstrating a focus on implementing and refining NAS algorithms.
artnasneural-architecture-searchstate-of-the-artmachine-learning
yashsmehta/personality-prediction
May 2020 - Jun 2022
Experiments for automated personality detection using Language Models and psycholinguistic features on various famous personality datasets including the Essays dataset (Big-Five)
Contributions:1 release, 2 reviews, 93 commits in 2 years 1 month
pytorchlanguage-modeldeep-learningessaysdataset
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.