Member of Technical Staff, Applied Research — Sieve

Location: San Francisco, CA (Onsite)
Compensation: $150,000 – $350,000 base + 0.05% – 0.4% equity
Visa Sponsorship: H-1B, O-1, OPT supported
Experience Level: 2+ years
Employment Type: Full-Time
Headcount: 4 open seats

Sieve is the only AI research company exclusively focused on video data infrastructure and video intelligence. Their thesis: high-quality video training data is the key bottleneck limiting progress in video AI, robotics, AR/VR, generative media, and agentic systems. Sieve is building the infrastructure and datasets to solve that bottleneck.

• Exabyte-scale video infrastructure
• Proprietary video understanding systems and internet-scale data processing
• Partners with top frontier AI labs and Fortune 100 enterprise customers
• Series A raised from Matrix Partners, Swift Ventures, Y Combinator, and AI Grant
• Team of approximately 12

Video represents 80% of internet traffic — and Sieve is building the foundational layer to advance video AI at scale.

This is a deeply technical applied research engineering role sitting between research and production engineering. The focus is building scalable video understanding infrastructure and applied AI systems — not pure academic research.

Core ownership includes:

• High-performance video understanding building blocks
• Large-scale video processing pipelines
• Computer vision, audio processing, and text processing workflows
• Applied model optimization and pre/post-processing optimization
• Parallelized inference systems and pipeline orchestration
• Inference performance optimization and selective model fine-tuning
• Customer-driven problem decomposition
• End-to-end applied AI product delivery

• 2+ years of computer vision or audio processing experience
• Strong Python engineering background
• Hands-on PyTorch or similar ML framework expertise
• Clean, maintainable production code
• Strong communication with customers and external stakeholders
• Passion for video and media technologies
• Product mindset — not research-only
• Ability to break customer problems into technical building blocks
• Willingness to work onsite in San Francisco

• Open source contributions or strong GitHub / portfolio projects
• Startup early-hire experience
• Applied research independence
• Video AI domain expertise
• Robotics experience
• High-performance inference optimization and pipeline architecture

Strong signals include backgrounds from Meta, NVIDIA, OpenAI, Google DeepMind, Google, Waymo, Intrinsic, and David AI, with direct computer vision experience, robotics exposure, and applied performance engineering depth.

• Applied computer vision engineers
• Video AI specialists
• Robotics perception engineers
• Audio AI engineers
• Research engineers who ship production systems
• Startup technical builders with optimization depth

• Candidates with less than 2 years of relevant experience
• Profiles without independent research or technical ownership
• Those without a top-tier technical background
• Candidates with weak video or computer vision relevance

1. Application review
2. Initial screen
3. Technical chat with CTO
4. Technical interview
5. CEO conversation
6. Onsite
7. Offer

• Role is fully onsite in San Francisco — please only apply if you can commit to this
• H-1B, O-1, and OPT visa sponsorship available

Shortlisted candidates will be contacted by David Joseph & Co., the recruiting partner managing this search on behalf of Sieve.