ABOUT ZAPDOS LABS

Zapdos Labs builds AI video agents for safer and more efficient industrial operations. Our systems connect to existing cameras to detect safety risks, monitor operational sequences, analyze worker and vehicle movement, and understand events across multiple cameras.

We combine contemporary multimodal models with precise computer vision so that every alert is accurate, relevant, and supported by video evidence.

THE ROLE

We are looking for a Senior Computer Vision Engineer to own core parts of our perception stack, from raw video and spatial calibration to tracking, event understanding, evaluation, and production deployment.

This is a hands-on role. You will build models and pipelines, design experiments, investigate failures in real footage, and turn research into systems that operate reliably on factory floors.

WHAT YOU WILL DO
• Design and productionize detection, segmentation, pose, tracking, re-identification, and activity-recognition systems
• Build reliable pipelines for continuous video from fixed cameras in challenging industrial environments




• Develop multi-camera tracking and event association across viewpoints, zones, and time
• Use camera calibration, scene geometry, trajectories, and 3D representations to reason about physical interactions
• Detect events such as worker-vehicle proximity, PPE non-compliance, restricted-zone entry, unsafe actions, and procedural deviations
• Combine learned models, vision-language models, temporal reasoning, geometry, and deterministic filters
• Translate customer SOPs and safety requirements into measurable perception tasks
• Define evaluation datasets and metrics for event accuracy, false-alert rate, latency, robustness, and operational usefulness
• Build testing and simulation workflows using digital twins before deployment to customer cameras
• Improve performance across changing lighting, occlusion, camera quality, site layouts, uniforms, equipment, and behavior
• Optimize inference for edge or cloud

📌 Senior Computer Vision Engineer (India)
🏢 Zapdos Labs
📍 India