Springfield, VA/St. Louis, MO - Salary Range 100k-185k (TS/SCI)

Job Brief

We are looking for highly energetic, TS-SCI cleared, IT professionals who are interested in joining our seasoned team of technologists. We are focused on driving significant enhancements in the efficiency and effectiveness of enterprise-wide IT operations and performing routine O&M for an IC agency. The program itself supports a broad range of IT enterprise services for end-users around the globe, with smaller project teams that focus on particular areas (i.e. Platform as a Service, Application Services, Cloud Computing, High Performance Computing, API Management, etc.).

Responsibilities

• Rapidly prototype containerized multimodal deep learning solutions and associated data pipelines to enable GeoAI capabilities for improving analytic workflows and addressing key intelligence questions.
• Implement State-of-the-Art (SOTA) Computer Vision (CV) and Vision Language Models (VLM) for conducting image retrieval, segmentation tasks, AI-assisted labeling, object detection, and visual question answering using geospatial datasets such as satellite and aerial imagery, full-motion video (FMV), ground photos, and OpenStreetMap.

Requirements

• 5+ years of relevant experience.
• Demonstrated experience applying transfer learning and knowledge distillation methodologies to fine-tune pre-trained foundation and computer vision models for segmentation and object detection tasks using satellite imagery.
• Professional experience building secure containerized Python applications including hardening, scanning, and automating builds using CI/CD pipelines.
• Experience using Python to query and retrieve imagery from S3 compliant APIs and perform common image preprocessing using libraries like Boto3 and NumPy.
• Experience with deep learning frameworks such as PyTorch or Tensorflow to optimize convolutional neural networks (CNN) for object detection or segmentation tasks.
• Experience with version control systems such as Gitlab.
• Experience leveraging CUDA for GPU accelerated computing.

Desired Qualifications and Skills

• Experience with the HuggingFace Transformers library and hub.
• Experience with OpenShift and container orchestration within Kubernetes using Helm, Kubectl, Kustomize, or Operators.
• Experience with Vision Transformers (ViT) such as DINO or DeiT.
• Experience communicating methodological choices and model results.
• Experience with verification and validation test benches.
• Experience with Explainable AI (XAI) techniques.
• Experience with Open Neural Net Exchange (ONNX).