Want to know if this job is worth applying to?
San Francisco, California, United States
Onsite
Mercor's mission is to organize human intelligence to power the AI economy. We're a leading AI data company, building the layer between human expertise and frontier models. Millions of domain experts on the platform are paid over $4 million per day to train frontier AI models. Mercor's APEX benchmark family measures AI's real-world impact on professional work. Mercor Enterprise brings this same infrastructure to Fortune 500 companies: helping companies capture how their best people actually work, translating that expertise directly back into agents.
Mercor is creating a new category of work where expertise powers AI advancement. Achieving this requires an ambitious, fast-paced and deeply committed team. You’ll work alongside researchers, operators, and AI companies at the forefront of shaping the systems that are redefining society. Mercor is a profitable Series C company valued at $10 billion. We work in-person five days a week in our San Francisco, NYC, or London offices.
Frontier AI is going physical, and the labs building it are bottlenecked on one thing: high-quality data from the real world. Mercor pairs its operational scale with the specialized engineering that physical-world data demands.
You'll build the data backbone of physical AI: the pipelines that take raw multi-sensor capture from the field (video, depth, inertial, audio, and more) and turn it into validated, privacy-safe, delivery-ready datasets for frontier labs. Ingest, segmentation, pre-labeling, automated QC, and packaging, at petabyte scale across thousands of concurrent collectors.
Build the end-to-end sensor data pipeline: ingest from capture devices in the field, through segmentation, pre-labeling, QC, and packaged delivery to customers
Design automated QC that validates recordings at scale: timing and sync integrity, calibration health, sensor continuity, coverage against requirements
Establish dataset schemas, versioning, provenance, and versioning, so every delivery has a clear system traceability
Build shared processing components such as privacy redaction, transcription, encoding, format packaging across all offerings
Integrate VLM-assisted pre-labeling and quality scoring into production workflows without sacrificing debuggability or human oversight
High ownership, early. This is a young, strategically central product area; the product you build will shape Mercor’s physical-world data collection standards
The data is the deliverable. The end product at Mercor is the data; what your pipeline produces is what shapes the models that large frontier lab trains on
Real physical-world scale. Your inputs come from devices operated by humans in global real world settings, for thousands of hours. Building systems that scale is precedent.
Strong production backend/data engineering experience — you've built and owned high-volume data pipelines
Experience processing video or sensor data at scale: large binary formats, streaming ingestion, distributed batch processing, object storage economics
Fluency in Python and comfortable with AWS
Genuine data taste: you can look at a sensor trace or a timing histogram and tell when something is off
Comfort in ambiguous, fast-moving problem spaces where requirements evolve with the customer
Experience with robotics data formats and tooling (MCAP, ROS bags, protobuf, Foxglove), camera geometry, or multi-sensor calibration and synchronization
Computer vision or multimodal ML experience (detection, tracking, VLM-based labeling or QC)
Prior work on data engines for AV, robotics, or egocentric video
Bi-annual performance bonus structure
Generous equity grant vested over 4 years
Up to $15k Relocation bonus
$10K housing bonus (if you live within 0.5 miles of our office)
$1.5K monthly stipend for meals
Free Equinox membership
$200 monthly laundry reimbursement
$200 monthly personal wellness reimbursement
Health, Dental, Vision insurance
