Senior Computer Vision AI Lead: YOLO FMV Pipelines

Socket.dev; Ashburn (VA); On-site; USD 104,000 - 218,000; Full time

Job summary

CACI seeks a Senior AI/ML Computer Vision Engineer to lead a 3-person pod, designing, training, and deploying custom YOLO models for CBP’s FMV feeds in Northern Virginia. You will build an end-to-end pipeline, define data curation standards, and guide mid-level engineers to achieve high FPS and strong mAP in a secure training/sandbox environment.

This role requires U.S. citizenship and a hybrid schedule in Ashburn, VA, with responsibility for Agile delivery and collaboration with

Qualifications

• U.S. Citizen with ability to pass a CBP background investigation.
• Education as stated: Bachelor’s, Master’s, or Ph.D. in Computer Science, Data Science, Electrical Engineering, or related AI-focused field.
• 7+ years of related technical experience, with 4+ years deploying DL/CV models (object detection).
• Local Ashburn, VA candidates must be able to work a hybrid schedule.

Responsibilities

• Lead a 3-person AI/ML engineering team, defining direction, sprint goals, code standards, and evaluation metrics.
• Design, modify, and train custom YOLO architectures for CBP FMV environments.
• Architect end-to-end pipeline to ingest and preprocess production video data in a secure training/sandbox.
• Define data curation and annotation guidelines; address class imbalance and challenging conditions.
• Optimize model inference speed (FPS) and accuracy (mAP) using CUDA, TensorRT, and related toolkits.
• Collaborate with Video Streaming Engineers, Software Engineers, and System Architects for integration.
• Serve as primary POC for the AI/ML pod in an Agile/Scrum setting.

Skills

Python

Leadership

Agile/Scrum

Deep learning

Computer vision

Model deployment

Video processing

NVIDIA CUDA

Education

Bachelor’s/Master’s/PhD in CS/DS/EE with AI focus

Tools

PyTorch

OpenCV

TensorRT

CUDA

cuDNN

TensorFlow

ONNX

CVAT

LabelStudio

Roboflow

Job description

CACI seeks a Senior AI/ML Computer Vision Engineer to lead a 3-person pod, designing, training, and deploying custom YOLO models for CBP’s FMV feeds in Northern Virginia. You will build an end-to-end pipeline, define data curation standards, and guide mid-level engineers to achieve high FPS and strong mAP in a secure training/sandbox environment.

This role requires U.S. citizenship and a hybrid schedule in Ashburn, VA, with responsibility for Agile delivery and collaboration with