Operations Engineer, Nimble Brain

Nimble Robotics•San Francisco, CA

8d•Onsite

About The Position

Nimble Brain turns real-world operational data into the training sets, evaluations, and feedback loops that make our robots smarter every day. That data comes from our collection sites — operators, supervisors, rigs, and capture systems running shift after shift — and every hour a rig is down, a calibration drifts, or a workflow jams is an hour of robot learning lost. As an Operations Engineer on Nimble Brain, you are the glue between operations and engineering. You'll be the technical backbone of our collection sites: diagnosing and resolving the issues that site technicians can't, translating floor pain into engineering priorities and engineering changes into floor-ready workflows, and constantly removing blockers before they cost a shift — then building the tools that prevent them from coming back. This role is equal parts triage and engineering — and most importantly, knowing which one a given issue calls for. Sometimes the right answer is the workaround that keeps a shift collecting; sometimes it's pulling a recurring problem out of the rotation and killing it for good with a tool, an interface, or a procedure anyone on the floor can run. Most times, it's both. And every fix carries a second judgment: is this one you build and ship yourself, AI tools in hand — or a true dev-team fix, and if so, whose? Quick action in the moment, long-term consequences in mind: stabilize now, industrialize what earns it, route what's bigger, and build the observability that makes tomorrow's issues smaller than today's.

Requirements

0–5 years of experience in a production environment supporting and improving complex robotics or hardware/software systems, or a minimum of 2 years of experience combined with a Bachelor of Science degree in Computer Science, Software Engineering, or Robotics.
Strong technical curiosity and debugging skills, including the ability to read technical logs and interpret data across the hardware/software boundary.
Scripting and data fluency — comfortable with Python, SQL, or equivalent, with AI tools as force multipliers, to analyze production data and build and ship lightweight internal tools yourself.
Demonstrated excellence in communication and documentation — this role's output is often a runbook, an interface, or a well-specified request, and all three have to be clear.
Ability to work autonomously with minimal supervision, across multiple sites and stakeholder groups.
Sound judgment under pressure: you can stabilize an issue in the moment without losing the long-term thread, and you know when a fix is yours to ship versus when it belongs to a dev team.
Comfortable hands-on with hardware: rigs, sensors, cameras, cabling, and the physical realities of a production floor.

Responsibilities

Act as first-line technical support beyond routine maintenance for our data collection sites: diagnose and resolve issues across rigs, teleoperation and capture systems, and the software that runs them.
Industrialize engineering workflows into floor-ready tools: build the operator-facing interfaces, checklists, and automations that let work once requiring an engineer run reliably in production — setup, calibration, diagnostics, and everything like them.
Build observability for the collection operation: rig health, calibration drift, capture pipeline status, and the alerts that surface problems before they cost accepted episodes.
Be the bridge between ops and engineering: own the routing call on every issue — fix it yourself, or spec it as a prioritized, well-defined request to the right engineering team — and translate engineering changes back into documentation, SOPs, and training the sites can absorb without losing a shift.
Partner with site technicians on the hardware line: they own routine maintenance and physical upkeep; you own systems-level diagnosis, tooling, and the escalation path into engineering.
Use a data-driven approach to prioritize: quantify which failures and workflow frictions cost the most accepted episodes, fix in that order, and revisit the workarounds you've shipped to decide which have earned a permanent fix.
Support site ramps and new-site bring-up: rig installation, station configuration, and getting new floors to steady state fast.
Create and maintain documentation and runbooks in collaboration with technicians, supervisors, and engineering — so sites self-serve more every month and escalate less.