OpenMined Featured Contributor: March 2024

Where are you based?

USA

What do you do (i.e. studying, working, etc.)?

I am a Computer Science PhD student at the University of Vermont.

What are your specialties (i.e. Python development, Javascript development,community organization, etc.)?

My   expertise   centers   around   privacy-preserving   machine   learning.   I’m   passionate   about creating   machine   learning   and   data   analytics   techniques   that   are   not   only   secure   and trustworthy but also user-friendly and scalable.

How and when did you originally come across OpenMined?

I initially discovered OpenMined through the Udacity course "Secure and Private AI" by Andrew Trask. It was during this course that I was first exposed to concepts such as differential privacy, federated learning, and secure multi-party computation, which ultimately played a pivotal role in shaping my career path and motivating me to pursue a Ph.D. in this area.

What was the first thing you started working on within OpenMined?

My first technical contribution to OpenMined was as a part of the DP team, where I collaborated with the amazing Ishan and Teo on the implementation of automatic differential privacy in PySyft.  In our work, we implemented various operations within PySyft that mimic numpy operations while incorporating differential privacy. Essentially, this involved introducing carefully calculated noise to the output.

And what are you working on now?

At the moment, my primary focus is on my role as a staff member within the OpenMined research team, where I conduct research, oversee the coordination of the monthly OpenMined reading group, and provide support in recruitment and other organizational endeavors.

What would you say to someone who wants to start contributing?

If you're eager to contribute, start by joining our Slack channel and connecting with our community leads, who will provide guidance on where to start based on your interests. For those looking to contribute to PySyft, the Padawan program is an excellent entry point, as it provides an in-depth exploration of how PySyft works. If research piques your interest, we encourage you to apply for a position on the OpenMined Research team. Please keep in mind that both programs recruit in cohorts, so stay updated on announcements in Slack. Additionally, if you prefer a more direct approach, you can also contribute by addressing issues on our
GitHub repository.

Please   recommend   one   interesting   book,   podcast   or   resource   to   the OpenMined community.

If you're looking to dive into the world of differential privacy, I highly recommend the interactive book   “Programming   Differential   Privacy”   by   Joseph   Near   and   Chike   Abuah.   It's  worth mentioning that one of the authors happens to be my advisor, so I may be a tad biased, but I genuinely consider it to be one of the top resources for gaining practical insights into this field.