Description
The role builds infrastructure for reproducible training and evaluation datasets from physical-operation capture contracts. It covers multimodal ingestion, temporal alignment, provenance, quality controls, schema evolution, dataset versioning, indexing, filtering, sampling, curation, replay, storage, compute, and reliable access for training and evaluation. The position requires a bachelor's degree or equivalent practical experience, strong Python and C++, Rust, Java, or Go skills, production pipeline experience, and direct experience with multimodal provenance and temporal relationships. It is a senior software engineering role based on-site in the San Francisco Bay Area with an annual base salary of $190,000–$240,000, health coverage, benefits, equity, and relocation support.
