Overview
From November 2024 to April 2026, Simpack-Edge LLC, the joint venture of CEdge and Simpack LLC, served as a subcontractor on a large-scale data labeling program for an Intelligence Community customer. The team produced high-quality labeled datasets used to train AI and machine learning models for national security missions.
The work required cleared personnel on short notice, hands-on management, and managed delivery support in a fast production environment. The joint venture kept quality high across several imagery and video sensor types and delivered every dataset on time against 90 to 120 day algorithm sprint cycles.
Scope of Work
Data Labeling and Annotation
Our analysts drew accurate bounding boxes and attribute labels around objects of interest such as vehicles, structures, and equipment, following the program’s ontologies and evolving labeling guidance. The work spanned electro-optical, synthetic aperture radar, full motion video, and horizontal motion imagery.
Quality Control
Every dataset passed through multi-level review before submission, combining human-in-the-loop checks with automated validation. Real-time feedback between annotators and quality reviewers kept accuracy above 90 percent against the customer’s labeling standards for the full period, and personnel on improvement plans moved above 92 percent.
Workforce Transition and Management
When the prime contractor’s circumstances changed, Simpack-Edge moved the existing St. Louis workforce into the joint venture with no disruption to operations or billing. We mirrored CEdge’s HR, payroll, and compliance policies, completed offers, benefits, security training, and timesheet setup without a gap, and secured favorable HSPD-12 Tier 1 adjudications for all personnel within 60 days. A dedicated management team stayed in place throughout.
Custom Integration
Beyond what the agreement required, we built an API integration and automated job that let the prime contractor pull labor and timekeeping data straight from our systems, improving their visibility and simplifying reporting.
Results
- Held labeling accuracy above 90 percent for the entire engagement
- Delivered 100 percent of datasets on time against sprint cycles of 90 to 120 days
- Moved all staff into the joint venture without operational or billing disruption
- Maintained full workforce continuity while the prime contractor reorganized
- Earned outstanding informal feedback on quality, program management, and flexibility