An open API service indexing awesome lists of open source software.

https://github.com/sarthakpati/sarthakpati

Machine Learning Engineer with a flair for C++.
https://github.com/sarthakpati/sarthakpati

federated-learning machine-learning

Last synced: 2 months ago
JSON representation

Machine Learning Engineer with a flair for C++.

Awesome Lists containing this project

README

        

Static Badge



[![Follow @sarthakpati.bsky.social](https://img.shields.io/badge/[email protected]?style=social&logo=bluesky)](https://bsky.app/profile/sarthakpati.bsky.social)

### Current Role
Software Architect & AI Researcher at [Indiana University's Division of Computational Pathology](https://medicine.iu.edu/pathology/research/computational-pathology), and Technical Lead at the [Medical Working Group of MLCommons](https://mlcommons.org/en/groups/research-medical/) working on designing solutions for privacy-focused AI in Healthcare.

I believe open software fosters better science, and thus have been involved in multiple open-source projects and their associated research studies, including the [Federated Tumor Segmentation (FeTS)](https://www.fets.ai/) platform and the [Cancer Imaging Phenomics Toolkit (CaPTk)](https://www.med.upenn.edu/cbica/captk/). I am currently focusing my efforts on the following:

- [Federated Learning for Postoperative Segmentation of Treated glioblastoma (FL-PoST)](https://fets-ai.github.io/FL-PoST/)
- [Generally Nuanced Deep Learning Framework (GaNDLF)](https://mlcommons.github.io/GaNDLF)
- [MedPerf](https://www.medperf.org)
- [Open Federated Learning (OpenFL)](https://github.com/securefederatedai/openfl) framework
- Creating and maintaining as many [Anaconda recipes](https://github.com/orgs/conda-forge/teams?query=%40sarthakpati) as possible to maximize reproducibility and scientific impact of open-source software

### Interests
- Applying concepts of AI (with a focus on privacy) to solve problems in healthcare.
- Committed to doing [reproducible](https://en.wikipedia.org/wiki/Reproducibility#Reproducible_research) and [deployable](https://en.wikipedia.org/wiki/Software_deployment) research.
- Firm believer of the saying _a weak algorithm that is well written & integrated is better than a strong algorithm that isn't_.
- Advocating for [F.A.I.R.](https://en.wikipedia.org/wiki/FAIR_data) in research.
- Helping people choose [the correct career path](https://oitecareersblog.od.nih.gov/2010/09/28/industry-vs-academia-which-is-right-for-you/).

### How to reach me
[patis [at] iu.edu](mailto:[email protected])