Machine Learning Engineer

Vantor

32d • $128,000 - $215,600

About The Position

Vantor is seeking a Machine Learning Engineer to build the systems that turn raw satellite imagery into trusted, well-labeled data, and to develop machine learning algorithms to turn that data into models. Your work sets the quality, speed, and traceability of the datasets behind Vantor's geospatial and machine learning products. You will work closely with internal domain experts, external annotation partners, and Vantor's Machine Learning and Software Engineering teams to design robust, scalable, low-latency solutions.
On the platform side, you will build pipelines for large geospatial imagery and metadata, including sensor, acquisition, and geolocation attributes, and move data reliably between annotation, storage, and model training systems. You will also track data lineage and versioning so every label can be traced back to its source image, guideline version, and annotator.
On the machine learning side, you will develop, adapt, and deploy models that generate proposed annotations, cutting manual labeling effort and turnaround time. You will build and operate production inference pipelines, evaluate model performance against labeled data, and use model feedback to decide which imagery should be labeled next. This is a hands-on role for an engineer who is as comfortable shipping a reliable data service as training and deploying a computer vision model, and who can connect the two into a single, continuously improving system.

Requirements

• 5+ years of relevant experience in machine learning engineering, data engineering, backend software engineering, or a closely related technical role.
• Hands-on experience training, fine-tuning, or adapting deep learning models using a modern ML framework.
• Experience deploying machine learning models into production, including building batch or real-time inference pipelines and managing model versions.
• Hands-on experience building and operating services on a major cloud platform, including managed compute, databases, storage, and identity and access management.
• Experience building and operating workflow orchestration for data or ML pipelines.
• Experience owning containerized services, CI/CD pipelines, and infrastructure-as-code.
• Proven ability to support production systems, including debugging, observability, secure configuration, and incident response.
• Strong Python application development skills.
• Bachelor's degree in Computer Science, Machine Learning, Data Engineering, Geospatial Science or GIS, Remote Sensing, or a related discipline, or equivalent demonstrated experience.

Nice To Haves

• Experience with Google Cloud Platform services, including Cloud Run, Cloud SQL, Cloud Storage, IAM, and service account management.
• Experience with Vertex AI, including Vertex AI Pipelines, training jobs, and model deployment, or with similar tools such as Kubeflow Pipelines.
• Experience with PostgreSQL and PostGIS, including spatial data modeling and managing schema migrations in production.
• Experience working with satellite or aerial imagery, including raster formats such as GeoTIFF, coordinate reference systems, and tools such as GDAL, Rasterio, Shapely, Fiona, and GeoPandas.
• Experience developing computer vision models for tasks such as object detection, semantic or instance segmentation, or change detection, ideally applied to overhead imagery.

Responsibilities

• Build systems that turn raw satellite imagery into trusted, well-labeled data.
• Develop machine learning algorithms to turn that data into models.
• Build pipelines for large geospatial imagery and metadata, including sensor, acquisition, and geolocation attributes.
• Move data reliably between annotation, storage, and model training systems.
• Track data lineage and versioning so every label can be traced back to its source image, guideline version, and annotator.
• Develop, adapt, and deploy models that generate proposed annotations, cutting manual labeling effort and turnaround time.
• Build and operate production inference pipelines.
• Evaluate model performance against labeled data.
• Use model feedback to decide which imagery should be labeled next.
• Ship a reliable data service.
• Train and deploy a computer vision model.
• Connect data services and computer vision models into a single, continuously improving system.

Benefits

• Robust 401(k) with company match
• Mental health resources
• Student loan repayment assistance
• Adoption reimbursement
• Pet insurance
• Incentive eligible with a target based on contribution, company performance, and/or individual results achieved

What This Job Offers

• Job Type: Full-time
• Career Level: Senior