Infrastructure Engineer

You'll build and run the infrastructure behind every machine we deploy: the software on the hardware, the pipeline that moves data from the field into model training, and the tools we use to manage all of our deployments.

• Build and maintain the software that runs on our deployed hardware: logging, telemetry, remote updates, and the services that keep the machine's computer healthy without anyone on site.
• Own the data pipeline end to end: getting logs off machines with unreliable connectivity, storing and organizing them, running labeling and processing, and feeding training jobs.
• Build our internal deployment management tools: which machine is running what, where, with which config, and whether it's healthy.
• Lead deployments on site: install, configure, validate, and hand off, and improve the process so the next one is faster.
• Set up and maintain the cloud and CI infrastructure that ties all of this together.

• Experience building and operating production infrastructure: data pipelines, deployment systems, fleet or device management, or similar.
• Python and at least one systems language, on Linux. Comfortable both on an embedded box over SSH and in a cloud console.
• Experience with the usual infrastructure stack: containers, CI/CD, object storage, databases, and monitoring.
• You've managed data for machine learning before, or are ready to learn how training workflows consume it.
• Willing to travel to jobsites regularly. Experience with field hardware or heavy equipment is a plus.