We're a team of engineers, researchers and operators building real-time vision and intelligence-fusion systems deployed where the stakes are highest. If you want your work in the field, not the backlog — talk to us.
Own the detection and tracking models that turn live camera feeds into decision-grade events. You'll take models from research to production on real deployments running thousands of cameras, on-prem and air-gapped.
What you'll do
-
Design, train and optimize detection, tracking and re-identification models for live video.
-
Push inference to run efficiently on edge and on-prem GPUs (TensorRT, ONNX, quantization).
-
Build evaluation pipelines that measure accuracy against real-world deployment data.
-
Work with deployment engineers to tune models for each customer's cameras and scenes.
What we look for
-
4+ years in computer vision / deep learning with production models shipped.
-
Strong Python and PyTorch; solid grasp of modern detection & tracking architectures.
-
Experience optimizing inference for real-time, resource-constrained environments.
-
Pragmatic about the gap between benchmark metrics and field performance.
