Part of the DoW SBIR/STTR FY27 Release 2 topic list. Browse all 83 topics by industry or service.
Topic DON27TZ02-NV012 from the DoW STTR 2027 BAA, Release 2. Pre-release now; proposals due January 6, 2027, 12:00 PM ET.
Quick answer
DON27TZ02-NV012 is a Department of the Navy STTR topic seeking an explainable, mission-specialized LLM that acts as a training and instructional assistant grounded in authoritative Navy manpower, personnel, training, and education (MPTE) documents. It is an STTR topic that starts at Phase I; as an STTR, it requires a research-institution partner. Proposals are due January 6, 2027, 12:00 PM ET.
Solicitation at a glance
- Topic number: DON27TZ02-NV012
- Official title: Mission-Specialized Large Language Models for Operational Training and Knowledge Integration
- Agency: Department of the Navy
- Program: STTR (DoW STTR 2027 BAA, Release 2)
- Award type: Phase I (see topic text for Phase II)
- Award amount: Phase I base up to $200,000 plus an option up to $115,000. Source: Navy Release 2 instructions in DSIP.
- CMMC level: Level 2 (Self)
- ITAR: No
- Technology areas: None
- Pre-release (direct TPOC contact allowed): November 4 to December 2, 2026, 8:00 AM ET
- Opens in DSIP: December 2, 2026, 8:00 AM ET
- Proposals due: January 6, 2027, 12:00 PM ET
- Official source: DSIP Topics app (search DON27TZ02-NV012)
BW&CO's take
What the Department of the Navy wants
As experienced personnel retire or transition, the Navy finds critical institutional knowledge difficult to preserve and transfer, and its training relies on static materials and fragmented systems. Commercial LLMs hallucinate on organization-specific content and often lack traceable references. The Navy wants an LLM framework, using fine-tuning and retrieval-augmented generation, that teaches mission topics from NETC curricula, qualification standards, MyNavy HR competency frameworks, and technical manuals, adapts to user role and experience, and cites sources for every output. Phase I delivers a feasibility report and Phase II plan. Phase II delivers a prototype platform with full MPTE corpus integration, DCWF and KSAT-aligned learning paths, and xAPI and IEEE 1484.20.3 interoperability.
Who is a good fit
- Retrieval-augmented generation and LLM fine-tuning work with measured hallucination reduction and source citation
- Adaptive instructional systems or intelligent tutoring products with competency-based learning paths
- Experience integrating with learning record stores via xAPI and competency standards such as IEEE 1484.20.3
- Deploying AI in regulated or zero-trust environments with documented data handling and model versioning
Watch-outs
- This is an STTR topic, so you need a formal research-institution partner performing a meaningful share of the work.
- Explainability is a critical requirement: users need transparent, auditable reasoning pathways and source citations for all generated outputs.
- Phase II requires user evaluation in representative mission environments measuring time-to-proficiency, knowledge retention, and user trust, plus a deployment package fit for Navy zero-trust data governance.
- The topic requires CMMC Level 2 (Self). It is not marked as ITAR-restricted in the topic metadata.
Think your company fits DON27TZ02-NV012? Talk with BW&CO before the January 6 deadline.
Plan Your DON27TZ02-NV012 Submission
Official topic text (verbatim from DSIP)
The text in this section is copied word for word from the official DSIP topic as published on November 4, 2026. DSIP is the official source and topics can be amended. Always confirm the current version at dodsbirsttr.mil before you submit. Our commentary above is BW&CO's interpretation, not Government guidance.
Title
Mission-Specialized Large Language Models for Operational Training and Knowledge Integration
Objective
Develop and demonstrate an explainable, adaptive super intelligence (SI) capability that enhances Navy workforce readiness, technical training, knowledge transfer, and personnel development through personalized instructional interaction, competency-based learning support, and integration of authoritative Navy workforce and training resources.
Description
The Department of the Navy faces increasing challenges associated with workforce modernization, technical skill development, personnel turnover, distributed operations, and the growing complexity of mission systems and operational environments. As experienced personnel retire or transition, critical institutional knowledge and operational expertise are frequently difficult to preserve, transfer, and scale across the workforce.
Simultaneously, Navy training and workforce development communities must support increasingly divergent learning requirements across accession, technical education, and qualification pipelines. Existing training approaches often rely on static instructional materials, fragmented knowledge systems, and limited mechanisms for adaptive or individualized instruction.
Commercial large language models (LLMs) demonstrate strong broad-domain reasoning capabilities but are fundamentally limited in their ability to reliably incorporate organization-specific knowledge without hallucination. They are unable to incorporate organization-specific terminology, mission procedures, technical constraints, and authoritative instructional content unique to mission environments. Additionally, these systems frequently generate responses without traceable references, limiting user trust and reducing suitability for technical learning and operational instruction. This gap creates a strong need for a specialized LLM system precisely tuned to an organization's corpus and operational requirements.
The Navy seeks development of a hyper-parameterized, explainable LLM framework designed specifically to support mission-focused training and knowledge transfer. The system should enable personnel to learn new technical and operational topics through an interactive interface. Outputs should be grounded in authoritative manpower, personnel, training, and education (MPTE) documentation while also leveraging the broad contextual understanding available from foundation LLMs. The proposed capability shall support accelerated workforce proficiency development, preservation of institutional knowledge, and adaptive learning across operational, technical, and mission-support communities while providing transparent, auditable, and trustworthy instructional guidance tailored to user roles, experience levels, and training objectives.
Explainability is a critical requirement for organization-specific LLMs operating in mission environments. The proposed solution must provide transparent and auditable reasoning pathways that allow users to understand how conclusions, recommendations, or generated outputs were derived.
The proposed capability should function as an intelligent training and instructional assistant capable of:
- Teaching mission-specific topics using authoritative organizational documents and instructional materials
- Combining proprietary knowledge with broad-domain contextual reasoning from foundational LLMs
- Guiding users through complex technical, operational, engineering, and/or procedural concepts
- Adapting explanations and instructional depth based on user role, expertise, and/or training objectives
- Providing explainable outputs with traceable references to source documentation
- Reducing hallucination through structured retrieval and grounded response generation
- Supporting continuous updates as training materials, policies, and mission guidance evolve
Phase I
Phase I will focus on assessing the feasibility of a concept for an explainable, mission-specialized training LLM built on sample Navy corpora. The effort will determine the technical feasibility of integrating organization-specific Navy MPTE corpora into a training-oriented conversational SI system. It will identify and characterize the types of Navy documents to be targeted, such as NETC curricula, qualification standards, MyNavy HR competency frameworks, and technical manuals, and define how each would be ingested and leveraged by the LLM. Building on this, the effort will define a concept for hyper-parameterized LLM fine-tuning and retrieval-augmented generation (RAG) architectures appropriate for Navy instructional environments. It will also develop and validate an initial approach for reducing hallucination through structured retrieval and grounded response generation. Required Phase I deliverables include a technical feasibility report and a Phase II development plan.
Phase II
Develop a prototype operational training and knowledge-transfer platform suitable for deployment within Navy organizations. The effort will develop, demonstrate, and validate a scalable, mission-specialized LLM platform that incorporates full Navy MPTE corpus integration, including MyNavy HR, NETC qualification pipelines, and relevant technical documentation repositories. It will also develop, demonstrate, and validate adaptive instructional capabilities that respond to user role, experience level, and training objective, with competency-based learning path support aligned to the Defense Civilian Workforce Framework (DCWF) and Navy Knowledge, Skills, Abilities, and Tasks (KSATs). To support user trust and oversight, the effort will develop and validate explainability and auditability mechanisms that give users traceable reasoning pathways and source citations for all generated outputs. It will produce a prototype deployment package suitable for integration within Navy zero-trust data governance and cybersecurity frameworks, including documentation of data handling, model versioning, and update procedures. User evaluation will be conducted in representative mission environments, measuring time-to-proficiency, knowledge retention, and user trust. The effort will also demonstrate interoperability with existing Navy training management systems and learning record stores using xAPI and the IEEE 1484.20.3 Reusable Competency Definition standard. Final deliverables include a validated prototype platform, comprehensive technical documentation, and a transition plan for government evaluation, potential program-of-record insertion, and commercialization.
Phase III Dual Use Applications
Support transition to NETC and associated subordinate training commands as an integrated SI instructional tool within Advana and/or Jupiter data management (or other Navy-specified system) and MyNavy HR ecosystem, supporting Navy-wide workforce readiness programs.
Potential expansion to other DoW components and the Intelligence Community is anticipated given the generalizability of the mission-specialized LLM architecture.
The commercial dual-use potential is significant. The underlying hyper-parameterized, explainable LLM framework is directly applicable to enterprise workforce training, knowledge management, and instructional SI markets across healthcare, finance, legal, and critical infrastructure sectors. Small businesses developing this technology will have a clear commercial pathway through SI-powered training platform products for regulated industries requiring auditable, organization-specific instructional SI.
References
Lewis, P., et al. “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks.” Advances in Neural Information Processing Systems, 33, 2020. https://arxiv.org/abs/2005.11401
Bommasani, R., et al. “On the Opportunities and Risks of Foundation Models.” Stanford CRFM Technical Report, 2021. https://arxiv.org/abs/2108.07258
“DoD Data, Analytics, and Artificial Intelligence Adoption Strategy.” U.S. Department of Defense, Office of the Chief Digital and Artificial Intelligence Officer, 2023. https://media.defense.gov/2023/nov/02/2003333300/-1/-1/1/dod_data_analytics_ai_adoption_strategy.pdf
“ MyNavy HR/Transformation.” U.S. Department of the Navy, Office of the Chief of Naval Operations, 2023. https://www.mynavyhr.navy.mil/About-MyNavy-HR/Transformation/
Wei, J., et al. “Chain-of-Thought Prompting Elicits Reasoning in Large Language Models.” Advances in Neural Information Processing Systems, 35, 2022. https://arxiv.org/abs/2201.11903
“IEEE Standard 1484.20.3: Standard for Reusable Competency Definitions (RCD).” Institute of Electrical and Electronics Engineers, 2022. [modified in 2025 to 1484.20.3a] https://standards.ieee.org/ieee/1484.20.3/10749
Keywords
<p>Large language model; LLM; mission-specialized SI; explainable artificial intelligence; retrieval-augmented generation; Navy workforce readiness; adaptive instructional system; competency-based learning; knowledge transfer; manpower, personnel, training, and education; MPTE; hallucination reduction; federated learning; zero-trust AI deployment</p>
Frequently asked questions
What is DON27TZ02-NV012?
It is a Department of the Navy STTR topic to develop and demonstrate an explainable, adaptive super intelligence capability that improves Navy workforce readiness, technical training, and knowledge transfer by integrating authoritative Navy workforce and training resources.
When is DON27TZ02-NV012 due?
Proposals are due January 6, 2027, 12:00 PM ET. DSIP pre-release began November 4, 2026, and direct contact with the topic authors is allowed only until the topic opens on December 2, 2026, 8:00 AM ET. After that, questions go through DSIP Topic Q&A.
Is this a Phase I or Direct-to-Phase II topic?
It is a Phase I STTR topic. Phase I assesses feasibility using sample Navy corpora and requires a technical feasibility report and a Phase II development plan. Phase II develops a prototype training and knowledge-transfer platform. As an STTR, it requires a research-institution partner.
Which Navy documents should the system ingest?
Phase I should identify and characterize target document types such as NETC curricula, qualification standards, MyNavy HR competency frameworks, and technical manuals, and define how each would be ingested. Phase II calls for full Navy MPTE corpus integration, including MyNavy HR, NETC qualification pipelines, and relevant technical documentation repositories.