San Francisco, CA · Hybrid · Reports to Head of Vision & AI
Voxel is building the future of Computer Vision and Machine Learning for operations, risk, and safety. We use computer vision and AI to enable existing security cameras to automatically detect hazards and high-risk activities, keep people safe and drive operational efficiencies. Our technology addresses the key cost drivers for workers’ compensation, general liability, and property damage, which cost US employers over $500 billion annually. Our customers include Fortune 500 companies across grocery, retail, manufacturing, food and beverage, logistics, and pharmaceutical distribution. We’ve passed $10M ARR with strong expansion revenue. Based in SF, backed by industry-leading VCs.
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 strong software engineer to own the data infrastructure required to train and evaluate our ML models. You’ll work with petabytes of data streaming from thousands of live customer cameras, and build pipelines to ingest, search, label, version and store the highest quality of data required to train state-of-the-art computer vision models. Your work directly shapes how our applied ML team measures and improves model quality. You'll set technical direction, write code, make architecture calls, and partner closely with applied CV, ML infra and Platform engineers.
Drive the roadmap for the data infrastructure that powers Voxel's vision and ML capabilities.
Build petabyte-scale pipelines for ingestion, search, labeling, versioning and storage of camera data.
Develop methods for data mining and automated data collection.
Partner with applied ML engineers on dataset quality for training and evaluation.
Collaborate with the Data Ops team (HITL) to design processes to measure and improve data quality.
Understand the data needs of Vision and AI engineers and design scalable infra solutions that support model improvement and vision capabilities.
Collaborate with the Platform team to store and retrieve data efficiently on the cloud.
4+ years of experience building and shipping large scale software solutions.
Working knowledge of ML training and evaluation. Understand what makes a good dataset, how to measure model quality, and how data quality affects model performance.
Strong Python. Comfortable across the stack: data mining, data labeling, storage and retrieval.
Track record of owning something end to end: building data products valuable to internal customers.
Bias toward shipping. You'd rather ship something good this week than something perfect next quarter.
Strong communication skills.
Experience with implementing data compliance & data governance solutions.
Deep understanding of data for computer vision - object detection, tracking, video understanding.
Familiarity with human in the loop data annotation, auto labeling.
Equity through Voxel’s Equity Incentive Plan
Total compensation includes base salary, annual bonus, and equity
Comprehensive health, dental, and vision insurance
Competitive paid parental leave
Unlimited PTO and flexible work arrangements
Daily meals in-office, team events, annual company onsite
What We Do
Voxel uses computer vision and AI to enable security cameras to automatically identify hazards and high-risk activities in real-time, keeping people safe and driving operational efficiencies. Our technology targets the key drivers for workers’ compensation, general liability, and property costs while providing full site visibility. The Voxel platform works by sending real-time notifications of safety violations and risky behaviors to on-site personnel and providing detailed reports with analysis of past incidents.
