About the role
We are looking for a passionate and curious individual to join our Vision AI team. You will work on real-world computer vision problems — from training object detection models to deploying them on local and cloud servers. This is a hands-on role ideal for freshers or candidates with up to 1 year of experience who are eager to build and learn fast.
Key responsibilities
▪Collect, annotate and preprocess image/video datasets from open sources like Kaggle, Roboflow etc.
▪Train and fine-tune YOLO (or other detection models) for custom object detection tasks
▪Work with RTSP streams for real-time video feed processing
▪Optimize model inference to achieve target FPS (30–60 FPS post-computation)
▪Evaluate model accuracy and minimize false positives in detection pipelines
▪Deploy models on local servers, cloud platforms, or Nvidia GPU-enabled systems
▪Build and present small-scale dummy/proof-of-concept projects demonstrating model capabilities
Required skills
▪Basic Python programming
▪YOLO (v5/v7/v8 or custom)
▪Image training & annotation
▪Vision AI fundamentals
▪RTSP link integration
▪Open datasets (Kaggle, Roboflow)
▪Model accuracy & algorithm basics
▪False positive reduction
▪FPS optimization (30–60 FPS)
▪Local / cloud / Nvidia GPU deployment
Good to have
▪Prior demo/personal project with any object detection or classification model
▪Familiarity with OpenCV, NumPy, or PyTorch/TensorFlow basics
▪Understanding of precision, recall, mAP metrics
Job Types: Full-time, Permanent, Fresher, Internship
Pay: ₹15,000.00 - ₹30,000.00 per month
Benefits:
Work Location: In person
