Machine Learning Engineer

PRADCO Inc. is building practical decision support from trail camera data, helping turn animal behavior signals into recommendation-ready hunting insights. In this onsite role in Massachusetts, you will help own the prediction machine learning lifecycle from tagged camera images through deer movement predictions and hunt location optimization.

What you’ll do

• Design and train object detection and classification models to identify deer presence, sex, age class, and antler characteristics in trail camera imagery (for example YOLOv8, RT-DETR, or similar).
• Build and maintain an end-to-end ML pipeline, including data ingestion from cloud storage, preprocessing, training on GPU clusters, evaluation, and deployment using Triton, TorchServe, or comparable tooling.
• Develop individual deer re-identification models using coat patterns and antler morphology to track specific animals across cameras and over time.
• Engineer features from vision outputs alongside environmental signals (weather, terrain, moon phase, rut calendar) to support downstream behavioral prediction models.
• Implement ML Ops practices with Mlflow or Weights & Biases for experiment tracking, model versioning, and staged deployments.
• Partner with Data Engineering to optimize data pipelines and work with a Wildlife Biologist advisor to validate model outputs against real-world deer behavior.
• Monitor production model performance and maintain retraining pipelines to respond to data drift across seasons.

What you bring

• 4+ years of machine learning engineering experience with demonstrated production deployments.
• Deep proficiency in PyTorch, plus strong preference for experience with Ultralytics/YOLO or similar detection frameworks.
• Solid understanding of CNN architectures, transfer learning, and domain adaptation.
• Experience deploying models at scale on GPU infrastructure using AWS SageMaker, GCP Vertex AI, or equivalent platforms.
• Proficiency in Python and familiarity with pipeline tooling such as Kafka or Airflow.
• Strong ML evaluation fundamentals, including confusion matrices, mAP, precision/recall tradeoffs, and the ability to diagnose failure modes.
• Familiarity with time-series prediction approaches such as LSTMs, Prophet, or XGBoost for temporal data.

Additional fit

• Experience with re-identification (RelD) or few-shot learning tasks.
• Prior work involving wildlife imagery, agricultural computer vision, or other low-contrast, occlusion-heavy domains.
• Experience with Microsoft Azure.
• Interest in the outdoors or hunting is a genuine plus, with domain empathy supporting better products.

Tools you’ll use

PyTorch, Ultralytics/YOLO, YOLOv8, RT-DETR, Triton, TorchServe, Mlflow, Weights & Biases, AWS SageMaker, GCP Vertex AI, Python, Kafka, Airflow, CNN architectures, LSTMs, Prophet, XGBoost