Lead AI/ML Engineer (Computer Vision)

CACI•Ashburn, VA

1d•Hybrid

About The Position

CACI is seeking a Senior AI/ML Computer Vision Engineer with Agile methodology experience to join their BEAGLE (Border Enforcement Applications for Government Leading-Edge Information Technology) Agile Solution Factory (ASF) Team, supporting the Customs and Border Protection (CBP) client in Northern Virginia. The role involves leading a newly formed Computer Vision team, a 3-person pod focused on designing, training, and deploying custom YOLO models from scratch. The team will process live Full-Motion Video (FMV) feeds to automatically detect and parse critical objects like roads, vehicles, people, and custom tactical assets. The Lead Engineer will architect the offline training pipeline, define data curation and annotation standards, and guide junior engineers in building high-performance, low-latency inference models.

Requirements

Must be a U.S. Citizen with the ability to pass a CBP background investigation.
Bachelor’s, Master’s, or Ph.D. in Computer Science, Data Science, Electrical Engineering, or a related field with an AI/ML focus.
7+ years of related technical experience.
4+ years of dedicated experience developing and deploying Deep Learning/Computer Vision models (with a heavy focus on object detection).
Experience with deep learning frameworks (PyTorch preferred, TensorFlow) and Computer Vision libraries (OpenCV).
Experience with advanced implementation and fine-tuning of the YOLO family of models (e.g., Ultralytics YOLOv8/v9/v10, Custom YOLO backbones).
Experience training deep learning models from scratch, including custom anchor box design, loss function modification, and hyperparameter tuning.
Experience with GPU-accelerated computing using NVIDIA CUDA, cuDNN, TensorRT, or ONNX.
Experience working with video streams (RTSP, HLS, files) and frame-by-frame processing.
Expert level programming in Python and writing clean, modular, and optimized code.
Experience with data curation, active learning, and using/managing annotation tools (e.g., CVAT, LabelStudio, Roboflow).

Nice To Haves

Experience with multi-object tracking (MOT) algorithms (e.g., DeepSORT, ByteTrack).
Familiarity with cloud-based AI infrastructure (AWS SageMaker, EC2 GPU instances).
Experience deploying models to edge devices (NVIDIA Jetson, tactical hardware).
Prior experience working with federal agency datasets or Customs and Border Protection (CBP).

Responsibilities

Lead a 3-person AI/ML engineering team, defining technical direction, sprint goals, code standards, and model evaluation metrics.
Design, modify, and train custom YOLO architectures from scratch, optimized for CBP's FMV feeds.
Architect the pipeline to ingest, clean, and pre-process production-recorded video data into a secure training/sandbox environment.
Define annotation/labeling guidelines, manage dataset curation, and implement strategies for challenges like class imbalance, varying weather, and low-contrast environments.
Optimize model inference speed (FPS) and accuracy (mAP, Precision, Recall) using hardware-acceleration toolkits.
Collaborate with Video Streaming Engineers, Software Engineers, and System Architects to integrate the live CV inference pipeline.
Act as the primary POC for the AI/ML pod, translating system requirements into technical tasking within an Agile/Scrum environment.