Sarthak Pati is a founder and AI safety researcher with 11+ years building trustworthy, end-to-end AI systems, now co-founding VerySafe.ai to attach cryptographic proof to every model run and enable policy-aware deployment of frontier LLMs. He has led large federated learning and medical imaging infrastructure projects at institutions like Indiana University and the University of Pennsylvania, translating cutting-edge research into robust, production-grade software and open-source frameworks. A recipient of over $9M in NIH-funded research with publications in Nature and IEEE, he blends academic rigor with practical DevOps and ML engineering—evidenced by contributions to projects like SLIDE and conda-forge CI tooling. Based in Miami, he also runs Vaiyu Solutions, advising enterprises on AI operationalization and data monetization, and is active in standards work through MLCommons. Less obvious: his background spans low-level C++/real-time systems to cryptographic attestation, enabling him to bridge hardware-focused performance work with governance and compliance for regulated AI deployment.
11 years of coding experience
11 years of employment as a software developer
Bachelor of Engineering Biomedical Engineering, Bachelor of Engineering Biomedical Engineering at Manipal Academy of Higher Education
Doctor of Philosophy - PhD Computer Science, Doctor of Philosophy - PhD Computer Science at Technical University of Munich
Kendriya Vidyalaya Sangathan
10th Maths Science, 10th Maths Science at St. Xavier's High School, BBSR
Stackoverflow
Stats
1,198reputation
179kreached
59answers
14questions
Badges
opencv
top-5%
Github Skills (28)
conda-forge10
github-ci10
c-language10
python10
machine-learning10
cicd10
deep-learning10
cprogramming-language10
githubaction-workflow10
operation9
mergetool9
opencv9
lint9
mergesort9
linter9
Programming languages (13)
C++CCMakeSWIGGoHTMLJupyter NotebookCommon Workflow Language
Codebase for "SLIDE : In Defense of Smart Algorithms over Hardware Acceleration for Large-Scale Deep Learning Systems"
Role in this project:
ML Engineer
Contributions:24 commits, 1 PR, 14 comments in 9 days
Contributions summary:Sarthak primarily focused on making minor adjustments to the code, specifically related to compilation and merging of branches. Their commits involved modifications to the `Node.h` file, likely addressing compilation issues and potentially affecting the overall structure of the project. The user also appears to be involved in integrating external contributions, as evidenced by merge commits from other branches. These changes suggest contributions related to the core functionality of the deep learning system.
A place to submit conda recipes before they become fully fledged conda-forge feedstocks
Role in this project:
DevOps Engineer
Contributions:46 reviews, 394 commits, 61 PRs in 1 year 9 months
Contributions summary:Sarthak's contributions center around maintaining and improving the `conda-forge/staged-recipes` repository's infrastructure. Their commits primarily involve modifying the `.github/workflows/scripts/linter.py` script, which suggests a focus on automated checks and CI/CD processes. These changes include updating the linter to incorporate new validation rules, specifically around recipe maintainer approvals, feedstock names, and multi-output recipe structures, ultimately enhancing the quality and standardization of recipes submitted to conda-forge.
placeconda-forgerecipescondasubmit
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.