Job Description

Our client is seeking an AI Engineer to design and deploy predictive models that enable intelligent home energy decisions: from forecasting comfort and cost to optimizing EV charging and demand response events.

Key Responsibilities

• Design, train, fine-tune, and deploy computer vision and multimodal models for image captioning, video description, object and person re-identification, and security event detection.
• Develop vision-language and embedding-based systems using architectures such as transformers, CLIP, BLIP, contrastive learning frameworks, and deep metric learning models.
• Build and maintain scalable data pipelines for image and video collection, annotation, curation, augmentation, and quality validation.
• Develop discriminative embedding models and tracking pipelines that enable cross-camera and cross-session identity continuity.
• Implement action recognition and behavior detection models to identify security-relevant activities and support customizable user-defined detection rules.
• Integrate AI models into edge devices and cloud services while optimizing for latency, accuracy, power consumption, and reliability.
• Collaborate with embedded, backend, frontend, product, and UX teams to deliver production-ready AI capabilities.
• Design and maintain evaluation frameworks using automated metrics and human review to measure accuracy, robustness, and real-world performance.
• Monitor deployed models for drift, bias, false positives, and system performance, and continuously iterate based on feedback and telemetry.
• Implement privacy-conscious and ethical AI practices including on-device inference, data minimization, and secure handling of sensitive visual data.

Required Qualifications

• Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, Data Science, Machine Learning, or a related field.
• Three or more years of hands-on experience building production machine learning or computer vision systems.
• Strong foundation in deep learning, computer vision, and machine learning including CNNs, transformers, metric learning, and representation learning.
• Experience working with image and video datasets, annotation workflows, and large-scale training pipelines.
• Proficiency in Python and experience using frameworks such as PyTorch or TensorFlow.
• Experience deploying models into production environments including cloud services and edge or embedded platforms.
• Strong problem-solving skills and ability to communicate technical concepts clearly to cross-functional stakeholders.

Preferred Qualifications

• Experience with multimodal models, vision-language systems, and retrieval-based architectures.
• Familiarity with vector databases and similarity search for embedding storage and retrieval.
• Experience optimizing models for real-time inference including quantization, pruning, and hardware acceleration.
• Knowledge of privacy-preserving machine learning techniques such as federated learning or differential privacy.
• Experience with multimedia processing pipelines and tools such as FFmpeg, GStreamer, CVAT, or Label Studio.
• Exposure to sensor fusion techniques combining vision with audio or other modalities.
• Background in accessibility, content moderation, or user experience driven AI applications.