Design, train, evaluate, and deploy machine learning models for recommendations, search, and content understanding. Lead migration from classical methods to PyTorch deep-learning models served on GPUs. Own offline evaluation and A/B experimentation, collaborate with ML, data, and backend engineers on production systems, and establish modeling standards. The role also involves adapting embedding and ranking models, optimizing large-scale recommendation systems, and exploring foundation models, multimodal representations, and advanced retrieval techniques.

Lead a team of four backend and ML engineers building machine-learning-driven ad delivery. Own the ML roadmap, production model-serving stack, real-time decisioning, feature pipelines, experimentation, monitoring, and backend ad services across web and mobile. Partner with product, analytics, and client engineering teams to balance revenue, retention, and inventory. Contribute hands-on to the codebase while mentoring engineers, managing performance, and guiding technical strategy.

Build and productionize ML and LLM-powered recommendation features serving millions of users. Own end-to-end systems spanning data pipelines, feature engineering, model training, real-time serving, APIs, monitoring, A/B testing, and distributed infrastructure. Collaborate with Product, Design, Engineering, and Platform teams while optimizing performance, cost, latency, reliability, and user experience.