AI/ML Computer Vision Engineer - Hybrid (Ashburn)

CACI International Inc; Ashburn (VA); Hybrid; USD 87,000 - 182,000; Full time

Job summary

CACI International Inc. seeks a Mid-Level AI/ML Computer Vision Engineer to join the BEAGLE ASF team in Northern Virginia. You will train and evaluate YOLO models, prepare training datasets from video, and annotate data while collaborating with video engineers in an Agile setting.

The role requires 3–5 years in software or data science with 2+ years in CV model work, Python, PyTorch/TensorFlow, and a hybrid Ashburn-based schedule.

Qualifications

• 3–5 years of software engineering or data science experience.
• 2+ years hands-on experience building/training computer vision models.
• U.S. citizen with ability to pass CBP background investigation.
• BS in CS/Data Science/Software Engineering or related field.
• Hybrid schedule availability in Ashburn, VA.

Responsibilities

• Train, fine-tune, and evaluate YOLO object-detection models on proprietary data.
• Process and format raw video into high-quality training datasets.
• Annotate and curate datasets; generate synthetic data when needed.
• Develop inference pipelines and export structured predictions (JSON/Protobuf).
• Generate performance reports and curves to identify model blind spots.
• Collaborate in an Agile/Scrum team to deliver production-ready code.

Skills

Python

NumPy

Pandas

Scikit-Learn

PyTorch or TensorFlow

YOLO (v8/v9)

OpenCV

FFmpeg

CVAT

LabelImg

Git

Benchmarking (mAP, FPS, latency)

Education

B.S. in CS/Data Science/Software Eng or related field

Tools

Docker

ONNX

TensorRT

Jira

Confluence

Git

Job description

CACI International Inc. seeks a Mid-Level AI/ML Computer Vision Engineer to join the BEAGLE ASF team in Northern Virginia. You will train and evaluate YOLO models, prepare training datasets from video, and annotate data while collaborating with video engineers in an Agile setting.

The role requires 3–5 years in software or data science with 2+ years in CV model work, Python, PyTorch/TensorFlow, and a hybrid Ashburn-based schedule.