Train, fine-tune, and deploy custom object detection and classification models
Design efficient edge-to-cloud ingestion pipelines optimized for periodic camera view monitoring (processing high-res image refreshes on a 5-minute basis, rather than handling continuous live video streams)
Build and maintain scalable, automated MLOps pipelines using AWS services (SageMaker, S3, Lambda, ECS/EKS, API Gateway) managed via Infrastructure as Code (Terraform or CDK)
Develop robust backend APIs (FastAPI/Node.js) and intuitive frontend dashboards (React/Vue) to serve model inferences, manage internal data annotation workflows, and display real-time performance metrics
Master complex data transformations and dimensional debugging—if an array's shape needs to be (3,3,2) instead of (3,2), you know exactly how to reshape and validate it without breaking the pipeline
Requirements
5+ years in software engineering, with at least 3 years dedicated to machine learning and computer vision in a production environment
Deep expertise in Python and major deep learning frameworks (TensorFlow, PyTorch)
Proven track record of deploying and orchestrating ML models on AWS, including managing CI/CD pipelines, model registries, and handling data drift
Ability to balance algorithmic complexity with cloud compute costs, network bandwidth limits, and system latency
Comfort working across the stack, from writing SQL queries and optimizing database performance to tweaking UI components for end-user applications
A Master's or Bachelor's degree in Electrical Engineering, Computer Science, or a related technical field is preferred
P.Eng or PMP certifications are a strong asset for managing complex technical project delivery