Oct 2016 – Present · 9 yrs 5 mos · New York, United States · On-site

• 2025 - Present:
• Technical Leader for Ads ML Infrastructure, powering Ads and Product Ranking at Meta's Scale.
• 2020 - 2024:
• Technical lead for ML training in the Meta's AI Infra org.
• Dedicated to building highly reliable and efficient Machine Learning systems, with a focus on proactively monitored infrastructure. This enables machine learning engineers to concentrate on advancing the state of the art and creating AI-powered experiences that delight end users.
• My work spans multiple layers of AI Infrastructure, from hardware deployment and PyTorch core development to front-end ML authoring and evaluation frameworks. I specialize in developing fault-tolerant and elastic distributed training systems, enabling production jobs to scale across thousands of machines.
• Spearheaded an organization-wide cultural shift towards enhancing the reliability and efficiency of GPUs, fostering an operational mindset across teams.
• 2016 - 2020:
• Systems Engineer Lead focused on Performance, reliability and efficiency of our mobile applications. Tackling problems such as cold start time, scroll performance, battery drain and on-device storage, through proactive monitoring, and data driven architectural changes.