Responsibilities

Develop models for detection, segmentation, tracking, OCR, action recognition, and video analytics.

Build video ingestion, annotation, augmentation, training, evaluation, and inference pipelines.

Optimise models for near-real-time execution across cloud, data-centre, and edge platforms.

Integrate model events, confidence, timestamps, and health status into operator monitoring workflows.

Manage false positives, missed detections, model drift, changing weather, lighting, and camera conditions.

Monitor production accuracy, latency, throughput, resource use, camera availability, and model health.

Requirements

Bachelor’s or Master’s in Computer Science, AI, Electrical Engineering, Mathematics, or a related discipline.

5+ years of production experience in computer vision or deep learning.

Strong Python and TensorFlow, PyTorch, or Keras skills.





Solid experience with OpenCV and detection, segmentation, tracking, OCR, or video analytics.

Experience with data preparation, model evaluation, deployment, monitoring, and MLOps.

Experience with real-time, multi-camera, edge, NVIDIA, or mission-critical systems is preferred.

Soft Skills

Strong analytical and experimental mindset.

Clear understanding of the operational impact of false or missed alerts.

Ability to communicate confidence, limitations, and model trade-offs.

Strong collaboration with domain, software, verification, and MLOps teams.

Disciplined approach to reproducibility and documentation.

Ownership from experimentation through production operation.

📌 Computer Vision Software Engineer (Hyderabad)
🏢 Staffington
📍 Hyderabad