Harsh Bhatia is a Member of Technical Staff in AI with 11 years of post-PhD experience building large-scale AI and HPC systems, currently based in San Jose. He designs and ships production-grade training infrastructure and workflows for LLMs, having led AI-driven multimodal pipelines and massive multiscale simulation workflows at national labs and industry. His background blends deep academic research in visualization and vector-field analysis with hands-on systems engineering—optimizing performance across Docker, Kubernetes, Slurm, NCCL, and PyTorch to cut model build times from days to hours. At Lawrence Livermore he led scientific ML and deployable recommender systems for computational biology and at Samsung he scaled virtual-human generative models. He is an active open-source contributor to spack/spack, adding and modernizing package definitions and Python3 support for complex scientific stacks. Combining rigorous theory with pragmatic engineering, he excels at turning computationally intensive research into reliable, high-performance production systems.
10 years of coding experience
13 years of employment as a software developer
Bachelor's Degree Information & Communication Technology, Bachelor's Degree Information & Communication Technology at Dhirubhai Ambani Institute of Information & Communication Technology
A flexible package manager that supports multiple versions, configurations, platforms, and compilers.
Role in this project:
Back-end Developer
Contributions:21 reviews, 13 commits, 16 PRs in 3 years 1 month
Contributions summary:Harsh contributed to the spack/spack repository by adding and updating package definitions. These contributions include adding new packages like `memsurfer`, `dssp`, `ibm-databroker`, `faiss`, `py-flask-cors`, `callflow`, and `py-pytaridx`, as well as updating existing packages like `vtk`, `cgal`, `py-msgpack`, `py-mock`, `py-griddataformats`, and `py-pyrsistent` with new versions. The user also made updates to support Python3 in the memsurfer package and fixed build issues.
Numerically robust detection of Critical Points in vector fields
Contributions:6 commits, 1 PR, 7 pushes in 5 years 11 months
criticalvectorfieldspoints
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.