Sieve is a San Francisco-based company building data infrastructure for AI, with an exclusive focus on video data. It combines exabyte-scale video infrastructure with novel video understanding techniques to develop datasets for video modeling. The company operates at significant scale, indexing billions of videos, images, audio clips, and interaction traces sourced from real-world, digital, and simulated environments.
The company's technical work spans video data infrastructure, video understanding, multimodal data processing, data indexing, annotation, scoring, embeddings, and human QA. Its products deliver high-quality data filtered and scored for semantics, rights, artifacts, and task quality. Sieve has achieved significant revenue and is at the Series A stage of funding.
Sieve has a team of 15 people working in person at its San Francisco headquarters. Engineers have direct ownership over projects end-to-end, with an emphasis on writing clean, maintainable code and moving quickly without creating brittle systems. The company values strong communication skills and a bias to action, alongside a deep interest in video, media technologies, and frontier AI applications.
