Job Description
CDIT is seeking a Mid-Level AI/ML Computer Vision Engineer to support the BEAGLE Agile Solution Factory (ASF) team serving U.S. Customs and Border Protection (CBP) in Northern Virginia.
You will be part of a focused three-person computer vision pod building custom object detection and classification models with the YOLO framework. The team processes live and recorded Full-Motion Video (FMV) feeds to automatically identify roads, vehicles, people, and terrain features. This is a hands-on role: you will bring raw production video into a secure training environment, curate datasets, train and tune models, and deploy high-performance inference pipelines.
Key Responsibilities
• Model Training & Experimentation: Train, fine-tune, and evaluate custom YOLO object detection models on proprietary datasets; experiment with hyperparameters, augmentations, and model sizes to balance speed and accuracy.
• Data Engineering & Ingestion: Process, clean, and format raw video recorded in production environments into high-quality training datasets.
• Dataset Annotation & Curation: Manage the annotation lifecycle, including labeling images and video frames, verifying annotation quality, and generating synthetic data when needed.
• Inference Pipeline Development: Write Python scripts that parse video frames, feed them into inference engines, and export structured prediction data (JSON/Protobuf) for detected objects.
• Model Evaluation: Produce performance reports, confusion matrices, and precision-recall curves to find model blind spots and drive continuous improvement.
• Agile Delivery: Partner with the Senior AI/ML Lead, video engineers, and developers in an Agile/Scrum environment to deliver production-ready code in short sprints.
Required Qualifications
• B.S. in Computer Science, Data Science, Software Engineering, or a related discipline.
• 5+ years of software engineering or data science experience, including 2+ years hands-on building and training computer vision models.
• Python programming with core ML libraries (NumPy, Pandas, Scikit-Learn).
• Hands-on experience training neural networks with PyTorch or TensorFlow.
• Experience implementing YOLO (e.g., YOLOv8/v9) for object detection, segmentation, or classification.
• Image and video processing experience with OpenCV or FFmpeg.
• Data curation and formatting, with experience using annotation platforms such as CVAT or LabelImg.
• Git version control and collaboration on shared codebases.
• Strong analytical skills with a focus on benchmarking model performance (mAP, FPS, inference latency).
• Willing to work on site 2-3 days per week in Ashburn, VA.
• U.S. Citizenship and ability to pass a CBP background investigation, which includes but is not limited to:
-3-year check for felony convictions
-1-year check for illegal drug use
-1-year check for misconduct such as theft or fraud
Desired Qualifications
• Containerized development environments (Docker).
• GPU execution environments and model export formats (ONNX, TensorRT).
• Multi-object tracking libraries.
• Agile tools such as Jira and Confluence.
This role offers direct, hands-on work on applied computer vision for a federal border security mission, within a small, high-impact team.
