ankush babbar | ML Infra Software Engineer
AI Startup
San Francisco, CAOn-siteFull-time$140k – $220k / year
About this role
Ankush Babbar
ML Infra SWE @ Stripe | Ex-Salesforce | CMU-SCS MSE | NSIT
Seattle, Washington, United States (US)
500 connections • 2,265 followers
About
I am a Backend and ML Infrastructure Engineer with over five years of experience designing and scaling large-scale distributed systems. My expertise spans Java/Python/Go based microservices, RESTful API development, and cloud-native architectures using Docker, Kubernetes, and AWS, as well as ML infrastructure and orchestration using tools like Flyte and Amazon SageMaker. At Stripe, I manage and optimize Flyte (ML Orchestration) infrastructure on AWS to streamline ML workflows. I've led initiatives to automate ML CI/CD pipelines, and build a unified ML platform that enhances ML developer productivity. My work includes developing distributed training solutions using Amazon SageMaker to accelerate large-scale model training. Previously at Salesforce, I engineered large-scale Redis control-plane services with Spring StateMachine and K8s operators, achieving 99.999% availability across 20K+ instances. I also built a real-time Notification Service (1M+ concurrent users, 200K+ notifications/sec) using Spring Boot, gRPC, and AWS. I hold a Masters of Software Engineering from Carnegie Mellon University and a Bachelors in Computer Engineering from NSIT. I am passionate about building robust, scalable backend and ML infrastructure systems.
Total Experience: 7 years and 4 months
Experience
ML Infra Software Engineer - Stripe(https://www.linkedin.com/company/stripe) (Current)
Jan 2025 - Present (1 year and 6 months) in Seattle, Washington, United States
- Manage and optimize Flyte in production, enabling ML engineering teams to efficiently train models by reducing operational overhead and simplifying orchestration.
- Build a unified ML Platform on top of Flyte, streamlining ML workflows by reducing boilerplate code, facilitating experimentation, and providing seamless integration with Stripe’s ML infrastr