Summary
David Mavashev is a Principal Software Engineer with 14+ years of experience building high-throughput Ads systems and machine learning pipelines that serve billions of requests daily. He specializes in ultra-low-latency inference, scalable streaming architectures, and unifying permissions and account models across large organizations, most recently leading agentic AI/LLM and Ads ML infrastructure at Amazon. A hands-on tech lead and architect, he consistently converts complex cross-functional requirements into production-ready, cost-effective systems while driving performance improvements (including multi-fold micro-optimizations). Known for mentoring engineers into leaders, he elevates teams through strong observability, testing, and automation practices. Based in New York, he blends deep Ads domain expertise with practical ML ops and distributed systems experience that accelerates product launches and long-term extensibility.
11 years of coding experience
15 years of employment as a software developer
B. Sc, Computer Sciences, B. Sc, Computer Sciences at Ben-Gurion University of the Negev
Russian, Hebrew, English