AI/ML Engineer

You are a highly skilled and motivated AI/ML Engineer specializing in Computer Vision & Un-Supervised Learning, eager to join MostEdge's growing team. Your primary role will involve building, optimizing, and deploying advanced video analytics solutions for smart surveillance applications, focusing on real-time detection, facial recognition, and activity analysis. **Key Responsibilities:** - **AI/ML Development & Computer Vision** - Design, train, and evaluate models for: - Face detection and recognition - Object/person detection and tracking - Intrusion and anomaly detection - Human activity or pose recognition/estimation - Work with models such as YOLOv8, DeepSORT, RetinaNet, Faster-RCNN, and InsightFace. - Perform data preprocessing, augmentation, and annotation using tools like LabelImg, CVAT, or custom pipelines. - **Surveillance System Integration** - Integrate computer vision models with live CCTV/RTSP streams for real-time analytics. - Develop components for motion detection, zone-based event alerts, person re-identification, and multi-camera coordination. - Optimize solutions for low-latency inference on edge devices (Jetson Nano, Xavier, Intel Movidius, Coral TPU). - **Model Optimization & Deployment** - Convert and optimize trained models using ONNX, TensorRT, or OpenVINO for real-time inference. - Build and deploy APIs using FastAPI, Flask, or TorchServe. - Package applications using Docker and orchestrate deployments with Kubernetes. - Automate model deployment workflows using CI/CD pipelines (GitHub Actions, Jenkins). - Monitor model performance in production using Prometheus, Grafana, and log management tools. - Manage model versioning, rollback strategies, and experiment tracking using MLflow or DVC. - **Collaboration & Documentation** - Work closely with backend developers, hardware engineers, and DevOps teams. - Maintain clear documentation of ML pipelines, training results, and deployment practices. - Stay current with emerging research and innovations in AI vision and MLOps. **Required Qualifications:** - Bachelors or masters degree in computer science, Artificial Intelligence, Data Science, or a related field. - 36 years of experience in AI/ML, with a strong portfolio in computer vision, Machine Learning. - Hands-on experience with: - Deep learning frameworks: PyTorch, TensorFlow - Image/video processing: OpenCV, NumPy - Detection and tracking frameworks: YOLOv8, DeepSORT, RetinaNet - Solid understanding of deep learning architectures (CNNs, Transformers, Siamese Networks). - Proven experience with real-time model deployment on cloud or edge environments. - Strong Python programming skills and familiarity with Git, REST APIs, and DevOps tools. **Preferred Qualifications:** - Experience with multi-camera synchronization and NVR/DVR systems. - Familiarity with ONVIF protocols and camera SDKs. - Experience deploying AI models on Jetson Nano/Xavier, Intel NCS2, or Coral Edge TPU. - Background in face recognition systems (e.g., InsightFace, FaceNet, Dlib). - Understanding of security protocols and compliance in surveillance systems. In this role at MostEdge, you will have the opportunity to work with cutting-edge technologies and collaborate with a talented team to deliver innovative solutions in the field of AI and computer vision. You are a highly skilled and motivated AI/ML Engineer specializing in Computer Vision & Un-Supervised Learning, eager to join MostEdge's growing team. Your primary role will involve building, optimizing, and deploying advanced video analytics solutions for smart surveillance applications, focusing on real-time detection, facial recognition, and activity analysis. **Key Responsibilities:** - **AI/ML Development & Computer Vision** - Design, train, and evaluate models for: - Face detection and recognition - Object/person detection and tracking - Intrusion and anomaly detection - Human activity or pose recognition/estimation - Work with models such as YOLOv8, DeepSORT, RetinaNet, Faster-RCNN, and InsightFace. - Perform data preprocessing, augmentation, and annotation using tools like LabelImg, CVAT, or custom pipelines. - **Surveillance System Integration** - Integrate computer vision models with live CCTV/RTSP streams for real-time analytics. - Develop components for motion detection, zone-based event alerts, person re-identification, and multi-camera coordination. - Optimize solutions for low-latency inference on edge devices (Jetson Nano, Xavier, Intel Movidius, Coral TPU). - **Model Optimization & Deployment** - Convert and optimize trained models using ONNX, TensorRT, or OpenVINO for real-time inference. - Build and deploy APIs using FastAPI, Flask, or TorchServe. - Package applications using Docker and orchestrate deployments with Kubernetes. - Automate model deployment workflows using CI/CD pipelines (GitHub Actions, Jenkins). - Monitor model performance