Machine Learning Engineer - Wildlife Vision & Prediction

EBSCO Industries Inc.; Ipswich (MA); On-site; USD 120,000 - 180,000; Full time

Job summary

Moultrie is seeking a Machine Learning Engineer to own the prediction ML lifecycle from tagged camera images to deer movement predictions. You will design, train, and deploy the prediction layer that turns behavioral data into actionable stand recommendations for hunters.

This is a high-impact, high-autonomy role shaping the platform's technical direction. Responsibilities include designing detection models, building end-to-end pipelines, re-identification of deer across cameras, and integrating

Qualifications

• 4+ years of ML engineering with production deployments.
• Strong PyTorch and CNN/transfer learning knowledge.
• Experience deploying models at scale on cloud GPUs.
• Familiarity with ML evaluation and time-series data.

Responsibilities

• Design and train object detection/classification models (YOLO/RT-DETR or similar).
• Build end-to-end ML pipeline: ingestion, preprocessing, training, eval, deployment.
• Develop re-identification methods for wildlife imagery across cameras and time.
• Create features from vision outputs and environmental data for downstream models.
• Integrate ML Ops tooling for experiment tracking and model versioning.
• Collaborate with Data Engineering and Wildlife Biologist to validate outputs.
• Monitor production performance and retrain to address data drift.

Skills

ML engineering

PyTorch

GPU deployment

Python

MLOps

Education

Bachelor's degree in CS/Math/ML

Tools

AWS SageMaker

GCP Vertex AI

Azure

Job description

Moultrie is seeking a Machine Learning Engineer to own the prediction ML lifecycle from tagged camera images to deer movement predictions. You will design, train, and deploy the prediction layer that turns behavioral data into actionable stand recommendations for hunters.

This is a high-impact, high-autonomy role shaping the platform's technical direction. Responsibilities include designing detection models, building end-to-end pipelines, re-identification of deer across cameras, and integrating