VetsLearn vetslearn.com

Curriculum outline

105 self-study lessons in 14 tracks. Minutes are planning estimates, not measured completion times.

Content version 2026.10-2 · 2026-10-09

Each line shows the lesson number, title and planned minutes (reading + practice). Lesson data reviewed 2026-10-03. Every lesson is free to read at vetslearn.com/curriculum/.

Foundations

7 lessons · 32 + 135 min (2 hr 47 min)

  1. 01What AI actually is10 + 15
  2. 02Five useful prompts and how to check them3 + 20
  3. 03Prompt like an Operator6 + 15
  4. 04When AI is wrong3 + 20
  5. 05AI for your transition3 + 25
  6. 06AI vocabulary for choosing a workflow4 + 20
  7. 07Where to go from here3 + 20

Intermediate

10 lessons · 165 + 1515 min (28 hr)

  1. 01Python for AI work16 + 30
  2. 02AI API requests and safe response handling14 + 30
  3. 03Data fundamentals18 + 30
  4. 04Prompt engineering at scale18 + 30
  5. 05Retrieval-augmented generation, from scratch20 + 45
  6. 06Vector databases and embeddings16 + 30
  7. 07Agents and tool use18 + 45
  8. 08Evaluation and observability16 + 45
  9. 09Cost, latency, and trade-offs15 + 30
  10. 10Capstone: ship something real14 + 1200

Expert

10 lessons · 169 + 2805 min (49 hr 34 min)

  1. 01ML and deep learning: enough to be dangerous18 + 45
  2. 02LLM internals: how transformers work18 + 45
  3. 03Fine-tuning, post-training, and alignment17 + 45
  4. 04Production deployment patterns and infrastructure18 + 45
  5. 05Monitoring, observability, and incident response17 + 45
  6. 06AI security: prompt injection, model risk, supply chain18 + 45
  7. 07Federal AI compliance: NIST AI RMF from the engineer's seat16 + 45
  8. 08ATO and FedRAMP for AI systems16 + 45
  9. 09Procurement from the engineer's chair15 + 45
  10. 10Capstone: production AI in a federal context16 + 2400

Annex A · Cleared Track

7 lessons · 102 + 190 min (4 hr 52 min)

  1. 01What cleared AI work actually means14 + 20
  2. 02OPSEC for AI tools15 + 20
  3. 03On-prem and air-gapped AI16 + 30
  4. 04The cleared AI role landscape15 + 30
  5. 05The certs that count14 + 30
  6. 06Talking about your work without violating15 + 30
  7. 07Where to go from here13 + 30

Annex B · Leadership Track

7 lessons · 106 + 200 min (5 hr 6 min)

  1. 01What you actually need to know about AI as a leader15 + 20
  2. 02Vendor evaluation that doesn't get gamed16 + 30
  3. 03Writing AI requirements into a contract16 + 30
  4. 04Risk management at the leader's altitude15 + 30
  5. 05The budget cycle and color of money15 + 30
  6. 06Building and managing an AI team15 + 30
  7. 07The first 180 days as an AI leader14 + 30

Academy · Orientation & Placement

3 lessons · 9 + 270 min (4 hr 39 min)

  1. O1Mission, roles, and the learning contract3 + 90
  2. O2Data handling and tool authorization gate3 + 90
  3. O3Diagnostic and safe workspace setup3 + 90

Academy · AI-Aware Operator

8 lessons · 24 + 960 min (16 hr 24 min)

  1. F1What modern AI systems do3 + 90
  2. F2Safe use before prompting3 + 90
  3. F3Prompt like an operator: Context, Task, Standards, Evidence3 + 120
  4. F4Verification, sources, and calibrated confidence3 + 120
  5. F5Research, summarization, and drafting3 + 120
  6. F6Decision support and human oversight3 + 120
  7. F7Tools, multimodal inputs, and agent boundaries3 + 120
  8. F8Foundations field exercise3 + 180

Academy · Applied AI Practitioner

8 lessons · 24 + 1740 min (29 hr 24 min)

  1. A1Workflow audit and use-case selection3 + 180
  2. A2Reusable CTSE workflows and quality standards3 + 180
  3. A3Data preparation, provenance, and records3 + 180
  4. A4Source-grounded research and retrieval3 + 240
  5. A5Documents, spreadsheets, and multimodal work3 + 240
  6. A6Workflow evaluation and operational metrics3 + 240
  7. A7Team rollout, change control, and incident handling3 + 240
  8. A8Applied workflow capstone3 + 240

Academy · AI Application Builder

12 lessons · 60 + 8460 min (142 hr)

  1. B1Command line, Git, Python, and environments5 + 475
  2. B2HTTP, JSON, APIs, secrets, and failure handling5 + 475
  3. B3Provider-neutral generation APIs5 + 475
  4. B4Data engineering for AI applications5 + 535
  5. B5Software quality, testing, and safe telemetry5 + 535
  6. B6Structured outputs and bounded tools5 + 475
  7. B7Retrieval baseline from first principles5 + 535
  8. B8Production RAG: authorization, hybrid search, and grounding5 + 715
  9. B9Evaluation design and release gates5 + 595
  10. B10Agents and tool use under control5 + 715
  11. B11Deployment, operations, and cost evidence5 + 535
  12. B12Builder capstone and technical defense5 + 2395

Academy · Production AI Engineer

12 lessons · 60 + 12060 min (202 hr)

  1. P1Machine learning foundations for system builders5 + 595
  2. P2Transformer and inference internals5 + 715
  3. P3Adaptation, fine-tuning, and post-training decisions5 + 715
  4. P4Production architecture and inference serving5 + 715
  5. P5AI threat modeling and adversarial testing5 + 835
  6. P6Identity, authorization, privacy, and data protection5 + 715
  7. P7Evaluation engineering and statistical evidence5 + 835
  8. P8Observability, safe logging, and incident response5 + 835
  9. P9Reliable workflows and agents5 + 835
  10. P10Performance, reliability, and resource governance5 + 715
  11. P11Secure supply chain and continuous delivery5 + 715
  12. P12Production capstone, incident exercise, and board5 + 3835

Academy · Federal AI Mission Practitioner

9 lessons · 55 + 6185 min (104 hr)

  1. G1Mission, information, and authorization boundaries6 + 354
  2. G2OPSEC and approved AI use6 + 354
  3. G3FedRAMP 20x, cloud impact levels, and authorization boundaries6 + 594
  4. G4RMF and the Authority to Operate6 + 594
  5. G5Continuous authorization and DevSecOps evidence6 + 534
  6. G6Federal AI governance and high-impact use6 + 594
  7. G7On-premises, disconnected, and open-weights operations6 + 594
  8. G8Federal acquisition artifacts for engineers6 + 534
  9. G9Federal specialization capstone and board7 + 2033

Academy · AI Program Leader

7 lessons · 44 + 3076 min (52 hr)

  1. L1Leader's model of AI capability and accountability7 + 353
  2. L2Use-case portfolio, value, and stop decisions6 + 414
  3. L3Vendor evaluation and mission bake-off6 + 474
  4. L4Requirements, test strategy, and acceptance6 + 474
  5. L5AI risk governance and incident leadership6 + 474
  6. L6Funding, lifecycle resources, and fiscal role discipline6 + 414
  7. L7Team, operating model, and 180-day leadership practicum7 + 473

Academy · AI Career Evidence

4 lessons · 25 + 1415 min (24 hr)

  1. T1Role families and evidence-based gap analysis6 + 294
  2. T2Resume, profile, and interview translation with OPSEC6 + 354
  3. T3Portfolio, technical narrative, and credential selection6 + 414
  4. T4Career campaign and evidence defense7 + 353

Workspace

1 lesson · 4 + 30 min (34 min)

  1. 01Set up a workspace your AI can navigate4 + 30

How self-review and quizzes work

  • Each lesson has a practice exercise. The learner reviews their own work against the lesson. No instructor sees it or grades it.
  • 42 of the 105 lessons have a knowledge check. It is open-book and checks recall. It is not a secure exam or an independent assessment.
  • Practice uses made-up or public examples. Learners should not paste classified, controlled or personal data into an AI tool.

What a completion record means

  • It means the learner marked the lesson complete. Completion is self-reported.
  • It does not mean a skill was tested, and it is not a certification.
  • With an optional free account, completions in these tracks are saved to the account: Foundations; Intermediate; Expert; Annex A · Cleared Track; Annex B · Leadership Track. Academy and Workspace progress stays in the learner’s own browser.
  • VetsLearn has no team reporting and no LMS integration. It does not send an organization any record about a learner.

Questions: [email protected]