Test Systems Engineer

About Tau

Power conversion is the bottleneck between the compute, energy, and mobility industries and the future they're trying to build. Most of the industry is still solving it with decades-old approaches. Tau is fixing that. We're a venture-backed hardtech company in Redwood City, building power conversion technology for data centers, EVs, and energy systems. Several products are already moving into production.

We hire people who own problems end-to-end: power electronics, embedded software, controls, test and validation, mechanical design. We solve from first principles, not legacy playbooks. We're past "promising technology." Customers are buying, volume is coming, and the people who join now set the standard everyone builds on after.

You think in systems, and you build for reality. You'd rather ship the simplest reliable answer to an engineering question than the most sophisticated one. You know the difference between complexity that earns its keep and complexity for its own sake.

You default to structure without needing to be told to. You see a recurring issue and build the fix that prevents it from recurring, not just the fix for today. You document because someone else needs to be able to run this without you in the room, not because you were asked to.

You have real judgment about risk. You know when "good enough" is actually good enough, and when it isn't, and you're comfortable being the person who says so even when it's inconvenient. High-voltage safety, interlocks, fault handling: you don't treat these as boxes to check, you treat them as the reason the job matters.

You want ownership over a domain that's core to how the whole engineering org performs, and you want to be measured on the capability you build, not just the hours you put in.

Innovation shows what's possible. Disciplined engineering creates products. Creative ideas matter only when they become reliable, manufacturable products that solve customer problems.

The lab is where every engineering claim gets checked against reality. If the data's untrustworthy, every downstream decision, design, program timeline, and customer commitment inherits that risk without anyone noticing until it's expensive.

This role owns that. You're responsible for the test infrastructure, the standards it runs on, and the judgment call on what's good enough to trust. You'll work across Mechanical, Power Electronics, Controls, Embedded Software, Manufacturing, and Program Management, and you'll have real authority to say a system isn't ready, a result isn't credible, or a process needs to change, and make that stick.

We want someone who biases toward structure: documented, repeatable, built to outlast any one person's memory of how it works. Not because process is the point, but because at 40-50 people and growing, tribal knowledge is a liability. At the same time, this isn't a role where you build the org chart and step back. You're in the lab. You're commissioning systems, debugging with the team at 6pm, mentoring technicians on measurement methods.

Player-coach, not dictator. You set the standard by being good enough at the work that people want to follow suit, not standards by decree.

• Own the dynamo meters, inverter test systems, environmental support systems, and lab infrastructure end to end
• Set the standards for what counts as commissioned, calibrated, and trustworthy, and hold the org to them
• Plan and execute capability upgrades with minimal disruption to active programs

• Translate engineering objectives into safe, repeatable test capability
• Integrate mechanical, electrical, controls, instrumentation, cooling, automation, and safety systems
• Own the safety philosophy: interlocks, e-stop logic, fault handling, operator protection
• Commission and validate systems before engineering release. Nothing goes live on your say alone until it's earned it

• Specify sensors, instrumentation, DAQ architecture
• Build measurement strategies that answer the actual engineering question, not just collect numbers
• Validate accuracy through calibration, uncertainty analysis, correlation, repeatability
• Be the person who can say, credibly, "trust this data" or "don't"

• Run preventive maintenance, calibration, troubleshooting, repair
• Own lab organization, documentation, safety, and operating procedure as an operating system, not a set of one-off fixes
• Eliminate recurring issues through durable corrective action, not repeated patching
• Support procurement and lab upkeep where it affects test capability

• Partner with Mechanical, Power Electronics, Embedded Software, Controls, Manufacturing, Program Management, suppliers, and customers
• Push back on requirements, timelines, or designs when the lab side can't safely or credibly support them
• Translate cross-functional input into test capability that…