Top Computer Vision Jobs in San Francisco Bay Area, CA

Build and deploy the robotics perception stack for a field-deployed public transit system. Responsibilities include developing perception, sensor fusion, object detection, localization, and reinforcement learning models; optimizing integrated hardware/software architecture; creating data collection pipelines; improving model performance; identifying software bottlenecks; and shaping the technical roadmap. The role works closely with the founders and requires up to 15% travel to customer sites.

Own the end-to-end computer vision platform for fine-grained item identification and pricing. Responsibilities include designing and fine-tuning vision models, defining evaluation and confidence standards, building labeling feedback loops, prioritizing coverage versus accuracy, and deploying reliable, low-latency serving infrastructure at scale. The role also communicates model capabilities and trade-offs to technical and non-technical stakeholders.

Develop and deploy applied ML and computer vision models to understand technical drawings and specifications. Work on multimodal reasoning across diagrams, text, and structured data, translate research into production within ~1 year, present prior research, and collaborate closely with founders to build customer-impacting solutions.

Leads the technical vision and architecture for perception software on unmanned platforms. Owns sensor fusion, localization, object tracking, environment modeling, production deployment, and performance measurement. Collaborates with autonomy and machine learning teams, mentors senior engineers, guides technical strategy, and supports hiring. The role requires extensive hands-on experience delivering perception systems for robotics, aerospace, defense, or unmanned vehicles, including radar and low-light sensing.

Develop and tune computer vision and machine learning models for 2D imagery and 3D point clouds. Projects focus on detecting urban and roadway features, including traffic lights, utility poles, curbs, and road markings. The intern will implement geometry-based algorithms, optimize perception models for real-time edge hardware, and apply machine learning to geospatial infrastructure problems while collaborating with experienced engineers.

Train and evaluate computer vision models for embedded deployment, build efficient C++ inference pipelines, and optimize latency, memory, power, and sustained performance. Integrate models with embedded Linux, test on representative hardware and data, and investigate errors across preprocessing, inference, and system integration. Lead candidates will set architecture and evaluation standards while guiding model and software development.

Join Pakket to develop computer vision systems for warehouses, deploying solutions and shaping both product direction and company goals. You'll work hands-on with real-world operations.

Annotate construction drawings using bounding boxes, polygons, and segmentation masks for computer vision models. Maintain consistent, accurate labels, clean datasets, identify ambiguous cases, and help improve annotation guidelines. Meet weekly quality and throughput targets while communicating data issues to ML engineers. The role requires focused repetitive work, precise instruction-following, independent remote work, and clear written English. It is a remote, 3–6 month contract with potential full-time conversion.

Own the architecture, quality, and technical direction of a production perception system for home security hardware. Develop robust computer vision and machine learning pipelines for multi-camera tracking, sensor fusion, object identity, and edge GPU inference. Define evaluation metrics, regression infrastructure, observability, and failure handling while optimizing accuracy, latency, and cost. Lead cross-functional technical decisions and mentor engineers across perception, hardware, cloud, and product teams.

Lead Glacier’s distributed computer vision engineering team, owning the CV roadmap, technical priorities, execution, hiring, mentoring, and cross-functional collaboration. Guide model, dataset, compute, evaluation, and production-performance decisions without serving as a day-to-day developer. Oversee labeling operations and coordinate with Product, Software, Operations, Manufacturing, and Field Engineering to deliver computer vision systems for recycling robots and customer sites.