Company Background
Specter's mission is to help 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.
Our co-founders Xerxes and Philip are passionate about empowering our partners in the fast approaching world of physical AI and robotics. We are a small, fast growing team who hail from Anduril, Tesla, Uber, and the U.S. Special Forces.
The Role
Specter is hiring a hands-on CV/ML camera test engineer to measure how camera and optics choices affect the performance of our AI models. Working with our Optics & Camera Engineering Lead, this individual will turn experiment plans into lab and field setups, run the campaigns, push the data through our CV/ML models and report what changed and why. The work is scientific at its core: reason from first principles about optics, sensor physics and illumination, form a hypothesis, design a controlled experiment to test it, and let the data decide. This individual will collaborate with camera, CV/ML, software, hardware and field teams to put data behind every camera decision.
Responsibilities:
Form testable hypotheses from first principles (optics, sensor physics, radiometry and illumination), and design controlled experiments with defined variables, controls, sample sizes and success criteria before collecting data.
Execute experiments comparing sensors, lenses, ISP settings, IR illumination, focus and window conditions, and measure their effect on detection, recognition and LPR performance at range.
Build and maintain lab and outdoor test setups, including targets at known distances, controlled lighting and low-light scenes, IR illumination and chamber runs across temperature.
Run field data-collection campaigns across sites, times of day, lighting and weather, keeping conditions, metadata and ground truth consistent.
Run captured video through existing CV/ML models and evaluation tools, and compute model metrics alongside image-quality measurements (MTF, SNR, distortion, pixels on target).
Script captures, camera settings, data pipelines and analysis in Python and shell on embedded Linux devices so every run is reproducible.
Isolate the cause of performance changes across optics, sensor, ISP, illumination and environment, using statistical analysis rather than anecdote, and write clear reports that state confidence, limitations and recommendations.
Support camera validation (focus stability, thermal, condensation, IR range) through EVT, DVT and PVT.
Qualifications:
Comfortable working in a fast-paced environment and being scrappy.
Bachelor's Degree in Electrical Engineering, Computer Engineering, Physics, Mechanical Engineering, Computer Science, or equivalent.
2+ years of experience in hardware test, validation or experimental work.
Strong Linux command line proficiency (required), including SSH, logs, file and process management and shell scripting on embedded devices.
Proficient in Python for scripting, data processing and analysis.
A scientific, first-principles mindset: you reason from the physics to predict results, design controlled experiments to test those predictions, and change your mind when the data says so.
Hands-on with test setups and lab equipment, and comfortable doing outdoor field work, including at night and in bad weather.
Strong grounding in statistics and experimental design, including hypothesis testing, confidence intervals, variance, sample sizing and trends across campaigns.
Exceptional organization and communication skills (both written and oral).
No CV/ML background required. You'll use our existing models and evaluation tools and learn the rest on the job.
Nice to Have
Experience with cameras, image sensors, lenses or image-quality testing (MTF, SNR, low light, IR).
Exposure to computer vision tools (OpenCV) or ML evaluation metrics such as precision and recall.
Experience with video tooling such as FFmpeg, GStreamer or V4L2.
Experience with thermal chambers, reliability or production test.
Background in physics, optics or a research lab environment.
Tool Proficiency
Linux command line and shell scripting (required).
Python with NumPy, pandas and plotting libraries.
Git for scripts and test procedures.
Imatest or equivalent image-quality tools (preferred).