At Iverto.ai, we are developing advanced AI-powered surveillance and video intelligence platforms designed to help industries monitor, analyze, and automate safety, security, and operational workflows in real-time.
We’re looking for a passionate AI Developer Intern with a strong interest in Computer Vision and hands-on experience with YOLO object detection models. This is an exciting opportunity to work on real-time video analysis solutions in real-world environments such as industrial zones, smart workplaces, and surveillance systems.
You’ll be part of a dynamic team, collaborating with product engineers and backend developers to build, test, and optimize models that scale across devices and deployment environments.
Responsibilities:
- Develop, train, fine-tune, and test YOLO-based object detection and tracking models (e.g., YOLOv5, YOLOv8).
- Perform dataset annotation, preprocessing, and augmentation using image/video data from surveillance feeds, drones, etc.
- Evaluate model performance using metrics like mAP, IoU, Precision, Recall, and FPS; implement model improvements for accuracy and speed.
- Optimize models for real-time performance using ONNX, TensorRT, quantization, and pruning techniques.
- Collaborate with backend/API teams to integrate AI models into scalable systems and production-ready APIs.
- Apply OpenCV and custom video/image processing logic to enhance detection pipelines.
- Stay updated with new research and tools in object detection, tracking, and computer vision architectures.
- Participate in cross-functional discussions to align models with real-world use cases (e.g., zone alerts, crowd detection, worker safety).
- Troubleshoot, debug, and improve AI pipelines based on test results and field feedback.
- Write clean, well-documented, and version-controlled code.
Requirements:
Technical Skills:
- Strong Python skills, especially in deep learning and CV applications.
- Experience working with YOLO models (training, transfer learning, and deployment).
- Hands-on with deep learning frameworks: PyTorch, TensorFlow, or Keras.
- Proficient in OpenCV and general image/video processing techniques.
- Familiarity with model optimization tools like ONNX, TensorRT, and GPU-based acceleration.
- Experience with image dataset handling, annotation tools like LabelImg, CVAT, and version control.
