AI Engineer, Computer Vision

About the Company

Mill is a waste prevention technology company reimagining what it means to eliminate waste, starting with food. We build smart systems and infrastructure for homes, businesses, and municipalities that transform food scraps from landfill-bound waste into valuable resources.

Responsibilities

• Build and manage the end-to-end ML training pipeline including data ingestion, ground truth generation, and annotation tooling
• Train and evaluate segmentation, classification, and mass-estimation models for the camera pipeline
• Build cloud-side evaluation harnesses to monitor edge model performance in the field
• Own MLOps including reproducible training, experiment tracking, and model versioning
• Export and validate models for deployment to edge devices via optimization and quantization
• Design and build LLM- and agent-powered product features to turn waste data into customer recommendations
• Systematically analyze failure cases to drive data and modeling decisions

Requirements

• Strong fundamentals in computer vision and deep learning (segmentation, detection, classification, tracking)
• Fluency with modern ML approaches including VLMs, LLMs, foundation models, and agentic systems
• Experience building ML training pipelines and data annotation systems at scale
• Experience designing rigorous ML model evaluation metrics and harnesses
• Proficiency with cloud ML infrastructure such as AWS
• Familiarity with cloud-to-edge model deployment
• Proficiency in Python, PyTorch, and OpenCV
• Strong MLOps familiarity on AWS infrastructure

Preferred Qualifications

• Experience with video understanding, temporal consistency, and tracking
• Experience with foundation models for data annotation
• Experience with MLOps tooling like Weights & Biases, MLflow, or SageMaker
• Experience shipping LLM or agent-powered features in consumer or B2B products
• Hardware or IoT product experience with embedded computer vision systems
• Google Cloud or Gemini experience