Senior ML engineer with 12+ years designing large-scale learning systems — from recommendation engines serving billions of daily predictions to production LLM pipelines. Passionate about turning research into real-world impact.
Production systems designed for reliability, throughput, and real-world impact — from infrastructure foundations to model-level innovations.
Designed and deployed a low-latency feature store handling 2M+ QPS across globally distributed data centers, cutting p99 inference latency by 42%.
Case StudyBuilt a scalable RLHF fine-tuning framework for domain-adapted LLMs, reducing alignment training cost by 60% while maintaining benchmark parity.
PaperLed architecture for a two-tower retrieval + reranking system serving 500M+ daily active users, improving CTR by 18% through contextual bandits.
Blog PostOpen-sourced an end-to-end model monitoring toolkit covering data drift, concept drift, and automated retraining triggers — adopted by 200+ teams.
GitHubSharing ideas at the intersection of systems engineering and machine learning research.
Open to advisory roles, research collaborations, conference talks, and senior engineering opportunities. Always happy to connect.