SBIR/STTR Research Award
Intelligenic Proposal for OSD254-P001
Award amount
$150,000Program and phase
SBIR · Phase IAward year
2025Award details
- Awarding agency
- Department of Defense
- Branch
- Office of the Secretary of Defense
- Recipient company
- Intelligenic, Inc
- Proposal award date
- 2025-08-26
- Contract end date
- 2026-02-27
- UEI
- DZGRYDV6EKL1
- Solicitation number
- 25.4
- Topic code
- OSD254-P001
Research abstract
Intelligenic will provide a Retrieval-Augmented Generation (RAG) based platform that will automate the software development lifecycle (SDLC) to help the Department of Defense (DoD) build its own RAG-based software solutions. Our platform automates the SDLC comprehensively including discovery, design, code generation, QA testing, and deployment. This process is typically very complex, manually intensive, and highly inefficient. Often many tools and data sources must be utilized or integrated to produce working software. Our platform provides a no/low code solution that will integrate with these tools and data sources to automate their operation significantly reducing the effort and expertise required to produce working software. Through the use of the Intelligenic platform, we can increase productivity by at least 50% while significantly reducing the effort and cost associated with building RAG-based solutions.áCore to the Intelligenic platform is Retrieval-Augmented Generation (RAG). It is an advanced technique in AI/ML that combines two powerful approaches: retrieval-based and generative models. Traditional generative models, such as GPT (Generative Pre-trained Transformers), produce responses based on a fixed set of learned knowledge. However, these models are limited by their training data and often suffer from knowledge gaps or inaccuracies when applied to dynamic environments.áRAG addresses these limitations by allowing the model to dynamically retrieve relevant information from external sources (e.g., databases, documents, intelligence reports, real-time data) before generating its output. The generative AI model then combines the retrieved data with its internal learned knowledge to produce a more accurate, contextually grounded, and up-to-date response. This hybrid approach significantly improves the reliability and relevance of AI-generated outputs, as the AI is no longer constrained to just its pre-existing knowledge but can adapt and respond to new, real-time information.áThrough our user-friendly, no/low-code interface, the DoD can create, modify, and deploy RAG-based AI solutions minimizing the need for extensive coding expertise. This allows soldiers or data analysts to rapidly prototype and deploy AI models, saving time and increasing flexibility.
Technology keywords
Not reported
Awards with the same topic code
Verified by an exact topic code match. This is not a text-similarity match.
- MyRadar Total AI Weather Forecasting Solutions
- AI for Airworthiness Certification
- FlightEdge: Real-time, Distributed AI/ML Model Training for Predictive Maintenance
- Dynamic, Adaptable Routing with Generative AI and LiDAR Data
- VOCAL: Voice Originality Cloning Across Languages