Job Title

Overview Dosti Global is developing a new SaaS product and we are looking for a Computer Vision & Machine Learning Engineer to design, build and help integrate the AI-powered visual monitoring. The role involves using cameras, drones, and mobile robotic systems to scan manufacturing environments and apply machine learning tools to detect safety issues, anomalies, and operational inefficiencies in real time. What You Will Build A system that can: Detect safety violations (e.g., machine out of safety boundary, equipment unguarded) Identify operational issues (e.g., lights ON during off-hours, doors not locked) Monitor machine activity and utilization Track workflow progress via visual comparison (before vs after) Generate alerts and insights for dashboards Core Responsibilities 1. Computer Vision System Development Build and train models for: Object detection (machines, people, safety zones) Anomaly detection (unexpected states/events) Activity recognition (machine running vs idle) Work with image/video data from cameras, drones, or mobile devices 2. Edge AI & Data Pipeline Deploy models on edge devices (e.g., NVIDIA Jetson Orin) Optimize models for real-time inference Design pipeline:  Camera/Drone → Edge Device → Cloud (Azure) → Dashboard 3. Robotics / Camera Integration Integrate with: Fixed industrial cameras Mobile cameras / handheld devices Drones (if applicable) Work with video streams (RTSP/WebRTC or similar) 4. Model Training & Optimization Train models using frameworks like TensorFlow / PyTorch Fine-tune models with real-world factory data Improve accuracy, latency, and robustness 5. System Integration Collaborate with backend/frontend teams Send processed insights to APIs / dashboards Enable alert systems (e.g., safety breach notifications) Required Skills & Experience 3+ years of experience in: Computer Vision / Machine Learning 2. Strong experience with: Python OpenCV TensorFlow or PyTorch 3. Experience with: Object detection models (YOLO, SSD, etc.) Image/video processing pipelines 4. Understanding of: Real-time inference systems Model deployment on edge devices Preferred (Nice to Have) Experience with NVIDIA Jetson (Orin / Xavier) Familiarity with robotics systems or drones Knowledge of RTSP streams / video pipelines Experience with Azure IoT / cloud integration Background in industrial automation / manufacturing environments Key Deliverables - AI models for: Safety detection Anomaly detection Activity monitoring - Real-time video processing pipeline - Edge deployment (Jetson or similar) - Integration with Factory Pulse dashboard Ideal Candidate Profile Strong problem solver with applied AI experience (not just theory) Has worked on real-world vision systems (CCTV, drones, robotics, etc.) Comfortable working across hardware + software boundaries Can take ownership of end-to-end AI pipeline