Tech Lead Manager, Vision & AI

Voxel•San Francisco, CA

2d•Hybrid

About The Position

Voxel’s perception system is the technical core of everything we ship. Our models detect human activity, equipment interactions, environmental hazards, and operational state in real time across thousands of cameras in manufacturing, logistics, retail, and pharmaceutical environments. Safety was our wedge; it proved our platform works. Now customers are pulling us into operations: equipment utilization, workflow compliance, process efficiency. Every new use case runs through the perception team. We're hiring a Tech Lead Manager to own the applied ML work that drives every new capability we ship. You'll set technical direction, write code, make architecture calls, and build a small team of engineers who take modern ML (VLMs, segmentation, tracking, temporal models) and turn it into things that work in real warehouses, on all kinds of cameras, for demanding customers. This is a hands-on role. You'll ship weekly, debug production problems, sit on customer calls, and own outcomes end to end. If you want a research org behind you, this isn't it. If you want to see your model running in 200 facilities weeks after you trained it, read on.

Requirements

5+ years building applied ML systems that actually shipped: production code, real users, consequences when it broke.
Track record of owning something end to end: from messy data and ambiguous specs to a model running in production.
Strong Python. Comfortable across the stack: training pipelines, eval, deployment, inference at scale.
Fluent with modern vision and multimodal methods (detection, segmentation, tracking, VLMs) and opinionated about when to reach for which.
Bias toward shipping. You'd rather ship something good this week than something perfect next quarter.
Comfortable with ambiguity, messy data, and customers who don't know what they want until they see it.
Some experience leading technical work: a project, a small team, a junior engineer you mentored. Doesn't have to be a formal management title.

Nice To Haves

Production experience with video understanding, temporal reasoning, or activity recognition.
Worked in a domain where failure has real cost (e.g. robotics, AVs, industrial, medical).
Built fast iteration loops: eval-in-the-loop, continuous improvement, deployment across many sites.

Responsibilities

Set technical direction for Voxel's vision and ML capabilities: what we build, how we build it, how we know it works.
Hire and lead a small team of applied ML engineers. Set the bar, set the goals, set the culture.
Stay in the code. Architecture decisions, model reviews, the occasional 2am debugging session.
Make hard calls on tradeoffs: classical CV vs. end-to-end models, on-device vs. cloud, accuracy vs. latency vs. cost.
Work directly with product, sales and customers. The loop from customer problem to shipped model runs through you.