• Take technical ownership of core components of recommendation and personalization systems (retrieval, ranking, evaluation). • Design and evolve two-tower / embedding-based retrieval models and downstream rankers. • Drive architectural and modeling decisions with a strong understanding of trade-offs between model quality, system complexity, latency, and cost. • Define and promote best practices for ML system design, experimentation, evaluation, and deployment. • Review ML designs, pipelines, and code with a focus on correctness, maintainability, and production readiness. • Act as a technical point of reference for ML-related decisions within the team. • Develop, train, and improve ML models for retrieval and ranking use cases. • Work with embedding-based deep learning models and classical ML approaches. • Perform in-depth data analysis, feature exploration, and systematic error analysis. • Build reproducible experiments and robust offline evaluation pipelines. • Optimize models for both offline metrics and online business KPIs. • Design and operate batch and real-time training and inference workflows in a cloud environment, with awareness of scalability and cost trade-offs. • Design, run, and analyze offline experiments and online A/B tests. • Own ML components in production, with a strong focus on reliability, observability, and safe iteration. • Monitor model performance and data quality in production. • Collaborate on scalable training and serving infrastructure for ML systems. • Participate in incident analysis related to ML systems and contribute to root-cause analysis and long-term fixes. • Design ML systems with failure modes in mind, including fallbacks and graceful degradation. • Work closely with Data Engineering on data pipelines and feature generation. • Partner with Product and Analytics to translate business goals into clear ML objectives and success metrics. • Act as a technical mentor for ML engineers, providing guidance on modeling, experimentation, and production ML. • Provide constructive feedback through code reviews and design discussions, supporting the growth of the team.