First seen by Alion on Nov 21, 2026.
Rao Information Technology is looking for a hands-on Computer Vision Engineer to work on AI Bot Eye, our video analytics product that detects fire, intrusion and other safety or operational events using existing CCTV cameras.
You'll develop computer vision features, improve detection accuracy and deploy applications on edge devices at customer locations. This role offers ownership from model development and application engineering through testing and field deployment. You'll join an existing product and codebase, with an initial technical handover.
Key responsibilities:
- Develop and maintain Python-based computer vision features, video-processing pipelines, and supporting backend and alert services.
- Prepare datasets, train or fine-tune detection models, and investigate false alarms and missed detections.
- Evaluate models on representative footage, tracking accuracy, detection delay and resource usage.
- Optimise inference performance across multiple camera streams, balancing CPU, GPU and memory consumption.
- Deploy, configure and troubleshoot applications on Ubuntu-based devices using Docker and system services.
- Diagnose camera-stream, network and application issues through logs, testing and remote access.
- Support technical demonstrations, customer installations and go-live checks, including occasional site visits.
- Maintain code in Git, document setup procedures and validate releases before deployment.
Required skills and experience:
- Strong practical Python skills, including reading existing codebases and debugging running applications.
- Hands-on experience with OpenCV and PyTorch or TensorFlow.
- Understanding of object detection, dataset preparation, model training and evaluation.
- Familiarity with precision, recall, confidence thresholds and false positives/negatives.
- Comfortable working with Ubuntu/Linux, terminal commands, SSH, permissions and application logs.
- Working knowledge of Git, Docker, APIs and basic networking, or demonstrated ability to learn quickly.
- Ability to investigate problems independently, verify solutions and communicate progress clearly.
- A relevant technical degree or equivalent practical experience.
Good to have:
- CCTV systems, RTSP/ONVIF, NVRs/DVRs or multi-camera video processing.
- NVIDIA Jetson, Raspberry Pi or GPU-enabled Ubuntu devices.
- ONNX, TensorRT, CUDA or inference optimisation.
- Experience deploying detection, tracking, face recognition or industrial vision applications.
What you can expect:
- Work on an existing AI product used with real customer CCTV systems.
- Own features from development through deployment and field validation.
- See the impact of your work on detection quality, application performance and customer operations.
Location and compensation:
This is a full-time, on-site role in Rajkot, Gujarat.
Ideally, you have 2-5 years of experience. Candidates with fewer years and strong relevant project work are also welcome.
Skills
OpenCV, PyTorch, Python, TensorFlow

