AI/ML Engineer (Computer Vision)

Description

The company is hiring a senior AI/ML computer vision engineer with Agile methodology experience for a team supporting U.S. Customs and Border Protection (CBP) in Northern Virginia, on a border enforcement applications program run through an Agile solution factory.

This is a full-time, regular role.

Responsibilities

• Lead a newly formed computer vision team of three engineers, including two mid-level engineers, setting technical direction, sprint goals, code standards and model evaluation metrics.
• Design, modify and train custom YOLO architectures from scratch, tuned for the resolutions, environments and challenges of CBP's Full-Motion Video (FMV) feeds.
• Process live FMV to detect and parse critical objects such as roads, vehicles, people and custom tactical assets.
• Design the offline training pipeline that ingests, cleans and pre-processes production-recorded video into a secure training and sandbox environment.
• Set annotation and labeling guidelines, manage dataset curation, and address class imbalance, changing weather and low-contrast scenes.
• Improve inference speed (FPS) and accuracy (mAP, Precision, Recall) using hardware acceleration tools such as NVIDIA TensorRT, DeepStream and CUDA.
• Work with video streaming engineers, software engineers and system architects to integrate the live computer vision inference pipeline into the wider application.
• Serve as the main point of contact for the AI/ML pod, turning system requirements into technical tasks within the Agile/Scrum process.

Requirements

• Required: U.S. citizenship and the ability to pass a CBP background investigation, which checks felony convictions over 3 years and illegal drug use and misconduct such as theft or fraud over 1 year, among other criteria.
• Required: A bachelor's, master's or Ph.D. in computer science, data science, electrical engineering or a related field with an AI/ML focus.
• Required: 7+ years of related technical experience, including 4+ years developing and deploying deep learning and computer vision models, with a strong focus on object detection.
• Required: For local candidates, the ability to work a hybrid schedule in Ashburn, VA.
• Required: Experience with deep learning frameworks (PyTorch preferred, TensorFlow) and the OpenCV library.
• Required: Advanced implementation and fine-tuning of YOLO models such as Ultralytics YOLOv8, v9 or v10, and custom YOLO backbones.
• Required: Training deep learning models from scratch, including custom anchor box design, loss function changes and hyperparameter tuning.
• Required: GPU-accelerated computing with NVIDIA CUDA, cuDNN, TensorRT or ONNX.
• Required: Working with video streams (RTSP, HLS, files) and frame-by-frame processing.
• Required: Expert-level Python and clean, modular, optimized code.
• Required: Data curation, active learning, and use or management of annotation tools such as CVAT, LabelStudio or Roboflow.
• Desired: Multi-object tracking (MOT) algorithms such as DeepSORT or ByteTrack.
• Desired: Cloud AI infrastructure, such as AWS SageMaker and EC2 GPU instances.
• Desired: Deploying models to edge devices such as NVIDIA Jetson or tactical hardware.
• Desired: Prior work with federal agency datasets or with CBP.

Conditions

Full-time, regular position.

Travel of up to 10%, local only.

No security clearance is needed to start.

Hybrid schedule in Ashburn, VA for local candidates.

Salary depends on factors including location, federal contract labor categories and wage rates, relevant experience, skills, education and certifications.

Proposed salary range: $103,800 - $218,100.

Equal-opportunity statement omitted.

Benefits

• Flexible time off that lets you take the time you need.
• Access to learning resources for career development.
• Comprehensive benefits covering healthcare, wellness, financial, retirement, family support, continuing education and time off.
• Competitive compensation.