Software Engineer, Robotics

Mercor•San Francisco, CA

1d•$130,000 - $500,000•Onsite

About The Position

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. As a Software Engineer on Robotics, you'll sit between Mercor's data systems and our customers' infrastructure. You'll work alongside operations and product teams to turn a set of requirements into a working delivery pipeline: the schema, fields, format, transport, and quality checks that prove what shipped is what was specified. This is a delivery-focused applied engineering role, not a classic robotics role. You will learn to grow and deeply understand robotics data types and work as part of a critical team leading the frontier development of physical AI, collaborating with engineers and researchers at frontier labs to ensure the data gets delivered properly.

Requirements

Strong backend and data engineering fundamentals in a modern language (Python, Go, Rust) and comfort operating production systems on AWS and GCP
Experience building and owning high-volume data pipelines, not just contributing to them
Experience with large binary formats, streaming ingestion, distributed batch processing, and object storage economics
Customer-facing instincts: you can lead a technical conversation, ask the right questions, and push back when a request is unreasonable
Comfort working through ambiguity and shipping iteratively with a product team, where requirements evolve with the customers and data types

Nice To Haves

Experience with robotics/AV data formats and tooling such as MCAP, ROS, protobuf, Foxglove
Prior work on data engines, pipelines, or delivery for multimodal data
Multi-sensor data experience across video, depth, inertial, and audio, including calibration and synchronization
Forward-deployed, solutions, or delivery engineering experience at a data or infrastructure company
Computer vision or multimodal ML exposure: detection, tracking, or VLM-based labeling and QC

Responsibilities

Partner with SPLs and client engineering teams to translate and implement data requirements into concrete delivery pipelines
Build and operate the pipelines that move data from Mercor's systems into customer storage reliably at petabyte scale
Format and transform datasets to per-client specification: MCAP and other container formats, custom schemas, field mappings, metadata, and versioning
Build automated validation to support delivery pipelines
Prototype quickly against new customer requirements, and debug issues when needed
Turn one-off client requests into reusable components, and own delivery infrastructure end to end