Specter · San Francisco
Hardware & ElectricalEngineeringFullTimeposted 4w ago
Your job is to build the machine that hunts down our hardware's weaknesses and gets that data in front of the people who can fix them. You're the hands-on end of it: test plans, fixtures, automation—and when the machine catches something, you chase it to root cause.
Specter’s mission is to help customers automate the physical world.
Today, we build video sensors with state-of-the-art AI agents that answer any question, anywhere in their environments. Our systems can automatically detect and reason about any physical activity captured on camera, from security incidents (e.g. perimeter intrusion, theft, LPR), to safety monitoring (e.g. PPE detection, injured people), to operational efficiency (e.g. material tracking, congestion monitoring). We offer both long range wireless (1km range) and wired sensor variants to suit any deployment.
Given a product and its requirements, you build the test plan, build the fixture, run the campaign, and script it so the next run is faster. When a unit fails—on the bench, in a chamber, or in the field—you own it across mechanical, electrical, and firmware boundaries until it's understood and fixed.
Concretely, here are some of the things you'll work on in the first few months:
- Plan and run campaigns — Turn product requirements into concrete test plans, then execute: set up fixtures, run chambers, stress hardware to its limits, collect the data, write the report.
- Automate — Script thermal, shock, and vibration campaigns in Python against instruments and fixtures so every run is reproducible.
- Root cause analysis — Chase failures to ground: tear down failed units, isolate the cause across hardware, firmware, and mechanical, and verify the fix. What you find inside reshapes the tests.
- Accelerated life testing — Run ALT and HALT to compress years of field wear into weeks, find the design margins, and feed end-of-life data back to the design team.
- Support EVT / DVT / PVT builds — Validation bring-up during NPI, plus reliability and abuse testing—thermal cycling, vibration, ingress, drops and mishandling, long-duration burn-in.
This is an early-stage team—your fingerprints will be on every fixture, script, and test report we have.
- 2+ years of hardware validation or test — Reliability, qualification, or systems test for embedded hardware products.
- Bench skills — Fluent with oscilloscopes, multimeters, logic analyzers, and power analyzers for bring-up and debug; thermal chambers and shakers are familiar territory too.
- Fixture hands — You've built or modified test fixtures and harnesses to get a unit connected, stressed, and measured.
- Reliability and stress testing — You've run mechanical, thermal, environmental, and abuse tests and can interpret what the results mean for the design.
- Structured root cause — You debug methodically across hardware, firmware, and system boundaries—teardowns included—and you don't declare victory on a hunch.
- Python and Linux — You drive instruments, sequence test runs, and make campaigns reproducible end to end—from a shell. Our products and rigs all run Linux; SSH and logs are home turf.
- Data rigor — You treat test data statistically: distributions, margins, trends across campaigns.
- Cross-functional by default — You work daily with EE, firmware, mechanical, and manufacturing without needing a process to make it happen.
- Production test systems — End-of-line stations, pogo-pin fixtures, or test that runs at production rate.
- EMC exposure — Pre-compliance testing or supporting formal certification.
- Manufacturing test — Production-line test experience or the validation-to-manufacturing handoff.