1 day ago
Sunnyvale, CA, USASenior
Responsibilities
- Build and scale data curation and enrichment pipelines that convert world-scale fleet data into high-signal training data.
- Develop mining, active-learning, model-based enrichment, and data-quality workflows operating at very large scale.
- Build offline, closed-loop, metric, benchmark, and world-model-based evaluation infrastructure for foundation models.
- Build and optimize distributed training, batched inference, serving, and large-scale model backfill infrastructure.
- Develop the data-platform backbone using distributed processing, vector search, lakehouse formats, dataset versioning, and enrichment and annotation catalogs.
- Turn one-off workflows into self-serve products with testing, observability, reliability, and on-call ownership.
- Partner with Applied Scientists and ML Engineers to move research systems from prototype to production.
- At TC4, set technical direction and drive ambiguous, cross-team platform problems to completion.
Requirements
- Strong production software engineering experience, especially with production Python, services, APIs, and large-scale data processing.
- Experience with large-scale data and distributed systems, including batch and streaming pipelines, workflow orchestration, and distributed processing.
- Experience designing reliable, observable, high-throughput data or ML systems, with strong SQL and query and performance optimization skills.
- Track record of shipping and operating production systems used by other teams, including testing, code review, observability, and on-call.
- Approximately six or more years of production experience for TC3, with more expected for TC4, or equivalent experience.
- Degree in computer science or comparable practical experience.
- Ability to own ambiguous cross-team problems and drive them to completion; TC4 candidates should be able to set technical direction and multiply team impact.
- Preferred exposure to ML platform or MLOps, model registration, distributed training, inference or serving optimization, foundation or world models, ML evaluation, vector search, annotation tooling, feature and data catalogs, Kubernetes, autonomous driving, robotics, or large-scale sensor-data workflows.
Tech Stack
Categories
Data EngineeringML Engineering
