Modular is the next-generation AI developer platform unifying the development and deployment of AI for the world.

The next generation of product breakthroughs will be powered by production quality infrastructure that brings together the best of compilers and runtimes, is designed for heterogeneous compute, edge to datacenter distribution, and is focused on usability. Unifying software and hardware with a "just works" approach that will save developers enormous time and increase their velocity.

We are incredibly excited about the mission before us, backed by some of the best investors in the world and if you are interested in joining us - just reach out via www.modular.com/careers.

Worked on the Einstein ML Platform - the platform for dozens of teams working on AI

applications at Salesforce.

I built the real-time ML Model Serving/ Inference solution. It’s a multi-tenant generic solution that supports a dozen of different applications such as Classification, Regression, Recommender, Natural language processing, Computer vision, etc. It serves any type of algorithms, from small linear models to the largest DL models, from models serialized to files to models packaged in containers.

It serves tens of thousands of ML models and millions of predictions per day.

I worked on Machine Learning platform for data scientists. Key projects:

• Serving layer for algorithms (1M RPS and 15M predictions per second)

• Feature store and A/B testing service for algorithms and services

• ML algorithms training and data pipeline and feature engineering

I led Cross-Device Audience Management project:

• Massive scale: 5B of user profiles, 8B of meaningful events per day

• High-load low-latency system: 1+M RPS <5ms

• Near real-time event-driven distributed architecture

I worked on a high-load Video on demand service (2M+ daily active users).