Staff AI Engineer

We're looking for a highly skilled and experienced Staff AI Engineer to bridge the gap between machine learning model building and real-world impact. You'll own the end-to-end deployment, monitoring, and optimization of our entire machine learning platform. A key focus will be deploying to edge devices, ensuring robust model health telemetry, and maximizing performance through hardware accelerators. If you're passionate about operationalizing ML models at scale and optimizing their performance in resource-constrained environments, you'll find an exciting challenge here at Sonatus.

• End-to-End CI/CD for ML: Design, build, and maintain robust Continuous Integration/Continuous Delivery (CI/CD) pipelines for machine learning models, ensuring automation, reliability, and reproducibility across the entire ML lifecycle from experimentation to production.
• Edge Deployment: Design, build, and maintain robust MLOps pipelines for deploying machine learning models to various edge devices (e.g., highly integrated into vehicle compute).
• Model Health Telemetry & Monitoring: Implement innovative monitoring and alerting systems to track model performance, data drift, concept drift, and overall model health in production environments. You'll also develop dashboards and reporting mechanisms to provide actionable insights.
• Model Optimization & Hardware Acceleration: Collaborate with ML researchers and hardware engineers to optimize models for performance, latency, and power consumption on specific hardware accelerators (e.g., GPUs, TPUs, NPUs, FPGAs).
• Infrastructure Management: Work with cloud platforms (AWS, Azure, GCP) and on-device environments to provision and manage the necessary infrastructure for model deployment and monitoring.
• Scalability & Reliability: Design and implement solutions for scalable and reliable model serving, ensuring high availability and low latency for inference.
• Data Pipelines: Design, build, and optimize robust data pipelines for machine learning model training, evaluation, and inference, ensuring data quality and availability.
• Troubleshooting & Debugging: Proactively identify and resolve issues related to model performance, deployment failures, and data discrepancies.
• Best Practices & Documentation: Establish and advocate for MLOps best practices, standards, and documentation to ensure efficient and consistent operations.
• Collaboration: Work closely with Machine Learning Engineers, Data Scientists, Software Engineers, and Product Managers to bring models from research to production.

• Minimum 7 years of work experience in MLOps, DevOps, or a similar role with a strong focus on machine learning systems.
• Proven experience deploying and managing ML models on edge devices (e.g., NVIDIA Jetson, Raspberry Pi, mobile platforms, custom embedded hardware).
• Deep understanding and hands-on experience with model health telemetry, monitoring, and alerting systems (e.g., Prometheus, Grafana, ELK Stack, custom dashboards).
• Strong experience with model optimization techniques for various hardware accelerators, including but not limited to NVIDIA Orin, NXP NPUs, GPUs, TPUs, NPUs, FPGAs (e.g., quantization, pruning, compilation for specific hardware, using frameworks like TensorRT, OpenVINO, TVM).
• Hands-on experience with popular ML frameworks such as PyTorch, TensorFlow, TFLite, and ONNX.
• Proficiency in programming languages including Python and C++.
• Solid understanding of machine learning concepts and the ML development lifecycle.
• Proficiency with containerization technologies (Docker, Kubernetes).
• Experience with cloud platforms (AWS, Azure, or GCP).
• Expertise in CI/CD principles and tools (e.g., Jenkins, GitLab CI/CD, Azure DevOps, GitHub Actions) applied to machine learning workflows.
• Excellent problem-solving skills and the ability to troubleshoot complex systems.
• Strong communication and collaboration skills with the ability to work effectively in a cross-functional team environment.
• Bachelor's or Master's degree in Computer Science, Electrical Engineering, or a related quantitative field.