Experience writing mission-critical code for production machine learning systems Experience building ML infrastructure, frameworks or services used by multiple teams Experience building or operating feature stores, or comparable ML data infrastructure serving production models Experience with embedding management: generating and versioning embeddings, refresh and retirement policy, and storing and serving them for retrieval at scale Solid understanding of the ML lifecycle: training, evaluation, deployment and serving/inferencing, with working experience building and deploying models Understanding of model evaluation, train-serve skew and data drift Working knowledge of deep learning architectures and training frameworks such as PyTorch or TensorFlow Prior experience applying ML at scale in advertising, recommender systems, information retrieval or related domains Experience with training data generation across multi-modal data (text, image and structured), including sampling and point-in-time correctness Experience building production data pipelines for ML systems where scale and performance are critical using distributed processing systems. Experience building ML systems using batch and streaming deployments, workflow orchestration and modern storage formats Strong data modeling and data architecture skills, with a high bar for system and data quality: correctness, reliability, testing and validation Strong problem solving, debugging and performance tuning skills, and pride in building automation, tooling and CI/CD Results oriented, with the ability to communicate effectively, both written and verbal, with technical and non-technical multi-functional teams Product-minded with a proven ability to seek projects with a sense of ownershipPreferred Qualifications.