Why I am Excited to Join Grid.ai as Head of Developer Advocacy
Today, I am excited to announce that after 7 years of Open Source Engineering and Developer Relations at Microsoft, I am joining Grid.ai to build its Developer Advocacy organization.
I look forward to the challenge and am thrilled to demonstrate what a passionate Developer Advocacy organization can do for both Grid and the PyTorch Lightning communities.
What is Grid.ai?
Grid.ai is an exciting startup founded by the creators of PyTorch Lightning, that aims to help machine learning engineers scale model training on their laptops without all the overhead of traditional MLOps.
In addition to the Grid platform, Grid.ai helps maintain the PyTorch Lighting project. Lighting is a lightweight PyTorch wrapper for high-performance AI research. PyTorch Lightning provides true flexibility by reducing the engineering boilerplate and resources required to implement state-of-the-art AI.
Organizing PyTorch code with Lightning, enables seamless training on multiple-GPUs, TPUs, CPUs and the use of difficult to implement best practices such as model sharding and even in 16-bit precision without changing your code.
Check out some of my favorite community projects powered by lightning below and feel free to reach out if you feel we’ve missed one.
Our bustling, friendly slack community has hundreds of experienced Deep Learning experts of all kinds and a channel for…
For those I haven’t had the pleasure of meeting yet, I’m originally from the Washington D.C area. My masters research is in the field of Multi Document Semantic Representation, Summarization and Coreference Resolution at the Bar Ilan Natural Language Processing lab and I additionally hold dual first degrees in History and Computer Science.
I am an avid AI enthusiast with a passion for history and engaging with new technologies. My areas of expertise are production AI/ML, NLP, CV, Python, Open Source and Web.
For the past 7 years I have been engaging with developer advocacy as an Open Source Engineer at Microsoft. As Microsoft has reinvented itself I had the opportunity to:
- Learn cutting edge 1st and 3rd party technologies. From Xbox, Office 365, Windows Phone and Hololens to the intricacies of Azure across countless services (VR, Modern Web, Containers, Machine Learning, Blockchain, IoT, Databases, Automation and more).
- Engage with non profits, mentoring students and young professionals from underrepresented high schools to the Ivy Leagues. Host and be hosted by top technical communities in Israel and abroad. Build keynote demos and speak at first and third party conferences across the world.
- Lead and contribute to business critical engagements with many of the top brands in the world. Collaborate with cutting edge start ups from the smallest seed companies to larger Series C and D organizations.
- Complete a Masters Degree at Israel’s top Natural Language Processing Lab. Publish AI Research at top Academic Conferences.
- Develop open source solutions and write content consumed by hundreds of thousands if not millions of developers around the world. Beta test and shape core Microsoft products from Windows 10 to Github and Azure ML.
Prior to joining Microsoft, I worked with the Inova Hospital System to build machine learning classifiers that detect references to new treatments in medical literature, with a healthcare Start Up, Secure Exchange Solutions, to automate HIPPA compliant emails, with Ben Gurion University’s Deutsch Telkom Research labs on the CEDRA II threat simulator, and at the National Institutes of Health to automate code for protein structure simulations.
I’m excited to take this experience and apply it to empowering the Grid.ai and PyTorch Lightning Communities.
Why am I Joining Grid.ai?
In both my industry work and academic research I have seen first hand the complexities of modern AI development. As Lightning Creator William Falcon pointed out,
“PyTorch is extremely easy to use to build complex AI models. But once the research gets complicated and things like multi-GPU training, 16-bit precision and TPU training get mixed in, users are likely to introduce bugs.”
There are plenty of supposed complex ML Ops workarounds for these challenges, but PyTorch Lightning is the first solution I’ve seen in the market that provides a practical and affordable answer this problem. Lightning structures your PyTorch code so it can abstract the details of training. This makes AI research scalable, easy to iterate and reduces the need for complex MLOps pipelines.
I’ll never forget the first time I scaled my first lightning project from one GPU to a cluster and everything just worked. After years of managing my own process ranks and versioning Horovod versions for every project, it was like magic. When I learned that I could get integrated logging through ML flow, Azure ML and tensorboard on all my metrics with without changing any code I was hooked.
Since then, everyone I have met on the Grid team has been modest, down to earth and extremely talented. A globally diverse team they all come from the some of the best institutions and companies in the world, while all having unique stories and perspectives that help make lightning stand out.
The AI market has the potential to transform society, and I believe that Lightning and Grid will be at the forefront of actualizing this dream through democratizing access to the state of the art. I’m excited to see what the future brings and I look forward to working with you all.
What will Developer Advocacy do for Grid.ai and the PyTorch Lightning Community?
The role of Developer Advocacy centers around the three tenants of :
- Community — you’ll see us both offline (whether it’s conferences, meetups, and user groups) and online (from forums to open source projects and social media outlets) meeting and collaborating with you. Our goal is to amplify and empower the voice of our community within the Grid.ai organization to make sure your needs are addressed.
- Content —Through engaging with the community we listen to you and then directly contribute your feedback to making our documentation and resources as empowering as possible. We convert your feedbacks into blog posts, articles, videos, as well as new open source projects and contributions based on your needs.
- Engineering — at the end of the day, we are all engineers. We connect with developers in the field, foster strong relationships with teams at Grid.ai and PyTorch lightning, and work together to improve the experience of building AI solutions making sure your voice is represented in the prioritization of our backlog.
My team will support you by doing things like:
- Developing open source code to unblock you and provide inspiration.
- Writing blog posts and articles about topics I believe will help you accomplish more.
- Ensuring you have the best possible documentation available.
- Developing strategic partnerships to simplify your code and take you models to the next level.
- Learning from you at user groups and conferences.
- Sharing learnings and updates with you at meetups and conferences.
- Taking your feedback back to the core product and engineering teams to make sure your needs are prioritized.
- Listening and growing, every day.
- Connecting with you over social media @pythiccoder.
If there is a feature or project you wish existed, feel free to let me know and if you work at another AI company and are interested in a partnering to help the AI community please reach out to me on LinkedIn.
About the Author
Aaron (Ari) Bornstein is an AI researcher with a passion for history, engaging with new technologies and computational medicine. As Head of Developer Advocacy at Grid.ai, he collaborates with the Machine Learning Community, to solve real world problems with game changing technologies that are then documented, open sourced, and shared with the rest of the world.