Video AI Engineer

Position: Video AI Engineer Client AT&T Location: Atlanta or Dallas Preferred/Not Required, Remote work for right candidate will be allowed Duration: 12+ Months Position Summary: Our client, a global leader in IoT solutions is seeking a highly skilled Senior Video AI Engineer with a minimum of 3-5 years of hands?on experience developing and deploying video AI models. The ideal candidate will be an expert in computer vision modeling, video management systems (VMS), and real?time inferencing, with proven skills in AI video preprocessing and data preparation. This role combines deep technical AI/ML expertise with practical knowledge of IT networking and video infrastructure to deliver production?ready solutions for our VaaS platform. The ideal candidate possesses a strong blend of hands-on technical experience in video solutions with strong technical acumen and strategic insight. Day-to-Day Tasks: AI Model Development & Optimization Develop, train, and optimize video AI models for object detection, classification, tracking, segmentation, and anomaly detection Implement transfer learning, hyperparameter tuning, and model fine-tuning techniques Optimize models for real-time inference using ONNX, TensorRT, OpenVINO, or similar frameworks Troubleshoot model performance, robustness, and deployment challenges Video Data Processing & Pipeline Management Design and implement large-scale video/image dataset preparation workflows Perform data cleaning, annotation, framing, resolution normalization, and noise reduction Video Management System (VMS) Integration Integrate AI solutions with VMS platforms including Milestone, Genetec, and Avigilon Work with streaming protocols (RTSP, WebRTC, RTMP) and video codecs (H.264, H.265) Deploy real-time video analytics for surveillance and security applications Configure AI models on camera systems (AXIS, Hanwha, etc.) Infrastructure & Deployment Deploy AI models in cloud environments (AWS, GCP, Azure) and edge devices Utilize containerization technologies (Docker, Kubernetes) and CI/CD pipelines Collaborate with DevOps teams to ensure scalable, secure deployment architectures Implement MLOps practices for model monitoring, retraining, and lifecycle management Systems Integration & Networking Configure enterprise network components including IP addressing, firewalls, and VPNs Troubleshoot system integrations to ensure seamless video and AI operation Required Skill Set: Modeling & AI Development Develop, train, and optimize video AI models for object detection, classification, tracking, segmentation, and anomaly detection. Work with tools such as/similar to Roboflow, Ultralytics (YOLO), CVAT, Supervisely, and TensorFlow/PyTorch pipelines. Perform transfer learning, hyperparameter tuning, and fine?tuning to adapt models to production environments. Optimize models for real?time inferencing using tools such as/similar to ONNX, TensorRT, or OpenVINO. Working knowledge and experience in Python AI Video Preprocessing / Data Preparation Experience in preparation of video and image data for AI modeling. Example includes: Cleaning and annotating datasets Framing, resolution adjustments, and aspect ratio normalization Noise reduction and filtering Segmentation and labeling for supervised training Ensuring compatibility with AI frameworks and deployment pipelines Work with large datasets to ensure high?quality training data for maximum model accuracy. Video & VMS Integration Experience with video management platforms (e.g., Milestone, Genetec, Avigilon) or integration with physical security systems. Integrate AI pipelines with Video Management Systems (VMS) for real?time analytics and monitoring. Example VMS system include Milestone, Genetec, etc. Handle RTSP, WebRTC, RTMP, and H.264/H.265 streaming for low?latency inference. Implement real?time video AI solutions in production VaaS environments. Working knowledge of video sent over a cellular and non-cellular network and video compression formats Deployment & Infrastructure Deploy AI models across cloud and edge environments. Experience in using Docker and Kubernetes desired but no required. Collaborate with DevOps teams for scalable and secure deployments. Leverage cloud services such as AWS/GCP/Azure services for model training, storage, and inferencing. Video over a network, Networking & IT Systems Configure and troubleshoot IP addressing, ports, firewalls, NAT, and VPN connections. Ensure smooth VMS and AI integration within enterprise IT infrastructure. Working knowledge of network security Camera systems Familiarity with camera systems – configuring, networking, deploying AI models in cameras. Example camera systems include AXIS, Hanwha etc. Architect and customer-facing skills Evaluate and recommend video analytic solutions to propose to customers based on requirements gathering. The ideal candidate possesses a strong blend of hands-on technical experience in video solutions with strong technical acumen, strategic insight, and consultative client-facing experience. Design, develop, and validate architectures utilizing video solutions that support real-time video ingestion, video analytics pipelines, object or event recognition, motion detection, and metadata extraction at the edge and in the cloud. Education: Bachelor's or Master's degree in Computer Science, Electrical Engineering, or related field