Harsh Vardhan
AI Research Scientist at Institute of software and Integrated System
Nashville-Davidson, Tennessee, United States
Summary
👤
Senior🎓
Top SchoolHarsh Vardhan is an AI research scientist with eight years of experience applying machine learning, reinforcement learning, and applied mathematics to data-driven model discovery, optimization, and reliable AI systems. He combines deep learning and Bayesian optimization to deliver sample-efficient training algorithms and AI surrogates that accelerate complex engineering workflows like CFD and FEA for autonomous systems and vehicle design. At Vanderbilt he developed anomaly and rare-event discovery methods that expose failures in RL controllers thousands of times faster than prior approaches and now contributes to DARPA’s CASTLE program building cyber-defense training toolkits. His background spans industry-scale production ML (Amazon supply-chain transformers, AWS/Spark pipelines) and field-deployable analytics in energy and IoT, reflecting both systems-level engineering and theoretical rigor. Based in Nashville with a PhD in Computer Science, he’s focused on trustworthy, cost- and energy-efficient AI — often squeezing substantial performance from far smaller datasets than typical deep learning pipelines.
8 years of coding experience
9 years of employment as a software developer
Bachelor's degree, Electrical and Electronics Engineering, Bachelor's degree, Electrical and Electronics Engineering at Shanmugha Arts, Science, Technology and Research AcademyDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at Vanderbilt University
Github Skills (11)
surrogate9
bayesian9
optimization7
energy7
deep-learning7
nist6
machine-learning4
pytorch4
deep-neural-networks4
data-generation3
scientific-machine-learning2
Programming languages (3)
Github contributions (5)
vardhah/UUV-design-optimization
Apr 2022 - Jan 2023
A design optimization study of underwater vehicle using Bayesian optimization and deep learning based surrogate model
Contributions:27 commits, 6 pushes in 9 months
surrogate-based-optimizationcfddeep-learningfinite-element-methodsoptimization
vardhah/Batch-mode-DeepAL-for-regression
Sep 2021 - Dec 2022
Contributions:47 commits, 5 pushes in 1 year 3 months
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.