Responsibilities

• Implement evaluation frameworks for Vision Language Models as automated assessors of Computer Vision model outputs
• Develop and test prompt engineering strategies and few-shot learning approaches to optimize VLM performance
• Create confusion matrix analyses and performance metrics to validate VLM accuracy against ground truth data
• Support iterative testing and optimization of model prompts, examples, and configurations
• Computer Vision & Image Processing:
  • Build systems to extract localized regions from large-format imagery (NIFT, PNG) based on CV model inferences
  • Implement configurable buffer zone parameters and metadata retention systems for localizations
  • Handle edge cases including overlapping detections and multi-scale object recognition 
  • Process imagery ranging from hundreds of MB to several GB in size
• Data Engineering & Pipeline Development:
  • Coordinate with model development teams to access and assess classified ground truth datasets
  • Support data access patterns using AWS services (S3, IAM) for classified data
  • Build ETL pipelines to process, transform, and manage large volumes of imagery and metadata
  • Ensure data quality, format standardization, and traceability throughout processing workflows

Requirements

• TS/SCI FSP Clearance
• DoD 8140 Compliance
• Technical Skills:
  • Experience with Vision Language Models (e.g. Claude, GPT-4V, LLaVA, CLIP) and prompt engineering
  • Hands-on experience with Computer Vision models and frameworks (YOLO, Detectron2, OpenCV)
  • Strong Python programming skills with experience in NumPy, Pandas, and image processing libraries
  • Experience with AWS cloud services (S3, SageMaker, Lambda, EC2)
  • Understanding of evaluation metrics, confusion matrices, and statistical validation methods
• Domain Knowledge:
  • Familiarity with geospatial imagery and remote sensing data
  • Understanding of Ground Sampling Distance (GSD), image resolution, and multi-scale analysis
  • Exposure to NITF format and geospatial metadata standards
  • Knowledge of few-shot learning and other prompting techniques
• Security & Environment:
  • Ability to work in classified/air-gapped environments
• Professional Skills:
  • Strong problem-solving abilities for complex technical challenges
  • Good communication skills to coordinate with model development teams and stakeholders
  • Ability to work collaboratively under the direction of senior engineers
  • Willingness to learn and adapt to new technologies and methodologies 
  • Experience documenting technical implementations and results

Desired Skills

• Bachelor's or Master's degree in Computer Science, Machine Learning, Computer Vision, or related field
• Experience with imagery analysis tools used in the Intelligence Community (iSpy, GeoServer)
• Previous work on large-scale search or detection problems across large datasets
• Familiarity with temporal analysis and change detection methodologies
• Experience with disturbed computing and optimization for large-scale processing
• Background in defense, intelligence, or national security applications

About Us
For more than 20 years, NewGen Technologies has solved our clients’ toughest IT challenges with integrity, security, and outstanding service by delivering both technology and talent. We have helped secure borders, have used artificial intelligence (AI) to fight terror, aided the identification of criminals, and have helped to prevent crime through the introduction of biometrics. Our team of Highly Cleared Specialists have hard-to-find skills and expertise in a wide spectrum of technologies to provide solutions that transform business processes and solve problems of national significance. #CJ