Applied Machine Learning Engineer

About the role

We are looking for an Applied Machine Learning Engineer to join our small but growing Machine Learning team. We use ML to improve the efficiency and accuracy of detecting and quantifying methane emissions, and we are actively expanding ML's role in our detection pipeline to reduce cost of goods, improve reliability, and enable the platform to scale to new geographies and customers. You’ll own production models end-to-end, from dataset and feature work through training, evaluation, and validation in production. You'll also help build the agentic AI systems we're developing for internal automation and customer-facing product capabilities.

What you'll do

• Train, iterate on, and improve the models in our detection pipeline, focusing on accuracy, efficiency, and generalization across geographies
• Build and automate training and retraining workflows with Dagster, and dataset and feature pipelines on top of our ML platform (ML flow, DVC)
• Design and run the offline experiments and evaluations that decide which model versions ship
• Build agentic AI systems that automate internal workflows and power customer-facing product capabilities
• Collaborate closely with our ML research partner on model development and our platform engineers on deployment, surfacing insights that shape ML platform and model priorities
• Build monitoring and observability into ML pipelines from the start, and share on-call responsibility for production ML systems

Qualifications

• Python proficiency and experience with at least one ML/DL framework (PyTorch preferred)
• 2+ years experience training models and building or operating ML pipelines in production
• Proficiency with Git and collaborative development workflows (branching, code review, CI/CD)
• Experience with SQL and relational databases (PostgreSQL preferred)
• Familiarity with data lake architectures and columnar storage formats (Parquet, S3)
• Familiarity with containerized deployments (Docker, Kubernetes)
• Experience with cloud computing providers, preferably AWS
• Comfortable working across multiple layers of the tech stack

Preferred Qualifications

• Experience with computer vision models and image datasets (familiarity with point cloud or LiDAR data is a plus)
• Experience with any of: KServe, MLflow, Dagster, DVC, or similar ML tooling
• Experience building LLM-based applications or agentic systems (tool use, evaluation, prompt engineering)
• Experience with geospatial data tools or extensions (PostGIS, GeoPandas, GDAL)
• Exposure to event-driven architectures (Kafka, CDC patterns)