Course
AI for Cloud Architects and Public-Sector Solution Engineers
Architecture, compliance, runbooks, cost models and the tools you build yourself.
Seriously hardTech support and IT
Start with
- AI at Work: FoundationsStart from zero and finish with three things you can actually use.
- AI for Tech Support and ITTriage, diagnose, script and document every ticket without pasting a single secret.
- AI for Managed IT ProvidersTwelve clients, one template: roadmaps, gap lists, runbooks and scripts with guardrails.
Who this is for
Cloud and solution architects, pre-sales engineers and senior infrastructure engineers serving regulated customers. Every example follows Fadi, an enterprise cloud architect on a public-sector team in Los Angeles, covering state and local agency migrations, compliance reviews and RFP responses.
You will be able to
- Produce an architecture document, a diagram as code and a decision log from discovery notes
- Map a solution to a compliance frame and answer an RFP section by section with sources
- Write a migration runbook and rehearse it against failure scenarios Claude plays
- Build a cost model with the levers the customer actually controls and a sensitivity table
- Stand up a proof-of-concept environment with Claude Code and infrastructure as code, with guardrails
- Connect Claude to internal documentation through MCP and score the assistant against a test set
Stops
- Guardrails firstThe four categories that never reach a chat, an account Project that holds no account data, and one write guardrail that governs every connector, script and CLI in this course. Set once
- Discovery to architectureTwo hours of discovery notes become an architecture document, a diagram written as code and a decision log where every choice carries the alternative it beat. Weekly
- Compliance mappingThe published control catalogue on one side, the design on the other, and a control-by-control table where every row states who implements it and which document says so. Weekly
- The RFP responseA compliance matrix built from the requirement list, section drafts written out of the account Project, a source check on every claim, and one rule about price that never bends. Weekly
- Migration runbookA cutover runbook generated from the architecture, rehearsed against failure injections Claude plays in role, with a rollback section that names the point of no return. Weekly
- Cost modelUsage assumptions written down and sourced, a spreadsheet built around the levers the agency actually controls, a sensitivity table, and every rate typed in by Fadi from the public calculator. Weekly
- Proof of concept with Claude CodeA written brief, Terraform generated in a repository and read line by line, a plan Fadi reads before any apply, a sandbox account with a hard boundary, and a teardown written before the build. Weekly
- MCP to internal docsA read-only connector to the team's reference library, an allow list of what it may reach, an explicit deny list of what it must never see, and a denial test set that proves the boundary holds. Weekly
- Evaluating the assistantThirty questions with expected answers, a rubric that scores sourcing and refusal as well as correctness, one graded run, and fixes that go into the Project and the connector rather than the next prompt. Weekly
- Fadi's quarter and the capstoneThree scheduled routines carry the quarter, and one agency goes end to end from discovery to proof of concept, graded out of a hundred against a rubric written into this stop. Weekly

Written for Claude. The ideas transfer to ChatGPT and Gemini; the menu paths don't. Current as of September 2026; where a feature may change we link to support.claude.com.
Claude helped Dasow draft and structure every course. A human at Dasow reviewed every stop.