AI Solution Architect: Discover, Estimate, Propose and Defend
3 short courses in 2 levels, 23 core lessons and about 6 hours. Follow one running project, an order and policy assistant, from the first workshop to the review board, and finish with a timed design on an unseen brief. There is no code: it is about decisions, documents and conversations. Pairs with the AI Architect course for the design theory and the Forward Deployed Engineer course for building.
The 3 short courses
Open one to see what you finish with and every lesson in it.
- Foundations about 1.5 h6 lessons
AI Solution Architect: Design Method and Patterns
Learn a seven-step method for turning a business problem into an AI design, and a catalog of eleven patterns to choose from and defend.
Not started
- Foundations about 1.8 h8 lessons
AI Solution Architect: Discovery, Scope and Security
Run an AI discovery workshop, set measurable outcomes and an MVP scope, and pass the customer's security, platform and compliance reviews.
Not started · 1 optional
- Advanced about 2.8 h15 lessons
AI Solution Architect: Estimate, Propose and Defend
Take an AI project from discovery to handover: briefing, estimate and TCO, build or buy, RFP, POC, review board, operations, and a blank-page defense.
Not started · 4 optional
AI Solution Architect: Design Method and Patterns
6 lessons (6 core) · about 1.5 h
A seven-step method for turning a business problem into an AI design, and a catalog of eleven patterns to choose from.
You finish with
- A seven-step map and a pattern shortlist for the order and policy assistant.
Lessons
- 1.Seven-Step AI Design Method on a Running Project17 minStart here
- 2.Patterns: Extract, Route and Explain13 min
- 3.Patterns: Retrieval, Hybrid Search and Documents15 min
- 4.Patterns: Act Safely and at Scale15 min
- 5.Integrating AI With Customer Order and Policy Systems15 min
- 6.Latency, Capacity and Availability Targets for AI Design15 min
AI Solution Architect: Discovery, Scope and Security
8 lessons (7 core, 1 optional) · about 1.8 h
Run a discovery workshop, set measurable outcomes and an MVP scope, and survive the customer's security, platform and compliance reviews.
You finish with
- A design pack: discovery notes, outcome sheet, scope, security checklist and rollout plan.
Lessons
- 1.AI Discovery Workshop: Questionnaire and Stakeholder Map15 minStart here
- 2.Turn a Vague AI Request into Measurable Outcomes15 min
- 3.MVP Scope and Acceptance Criteria for an AI Pilot15 min
- 4.Data Readiness Check for an AI Assistant15 min
- 5.Customer Security and Platform Reviews for AI Projects15 min
- 6.AI Demos and Change Control for Customer Projects15 min
- 7.AI Pilot Rollout Plan and Customer Solution Proposal15 min
- 8.Compliance Questions in Customer AI ReviewsOptional16 min
AI Solution Architect: Estimate, Propose and Defend
15 lessons (10 core, 4 optional) · about 2.8 h
From discovery to handover: an executive briefing, an estimate with TCO, build or buy, an RFP, a POC, a review board, operations, and a timed blank-page defense.
You finish with
- An estimate with a three-year cost table, a decision matrix, a POC plan and a review defence.
- A scored blank-page design and a 90-day practice plan.
Lessons
- 1.Solution Architect Cycle: Discovery to Handover12 minStart here
- 2.Executive Architecture Briefing for AI Projects15 min
- 3.AI Project Estimation: Effort, Tokens and TCO20 min
- 4.Build, Buy or Hybrid: AI Vendor Selection Matrix15 min
- 5.RFP, RFI and Statement of Work for AI Projects10 min
- 6.AI POC Design and Handling Objections9 min
- 7.Architecture Review Board Simulation for AI9 min
- 8.Operate and Hand Over an AI Assistant After Go-Live15 min
- 9.Blank-Page AI Design Defense on an Unseen Brief36 min
- 10.90-Day Practice Plan for Enterprise AI Design24 min
- 11.Solution Design Interview Questions: Sizing, TCO, RiskQuestion bank10 min
- 12.Lab: Support-Ticket Triage Assistant DesignOptional35 min
- 13.Lab: Invoice and Claims Document Processing DesignOptional35 min
- 14.Lab: Internal Knowledge Assistant With Access ControlOptional35 min
- 15.Lab: Agentic Order-Exception Handler DesignOptional35 min
AI Solution Architect Foundations: Discover and Design
A design method, a pattern catalog, and discovery, scope and security reviews on one project. About 3.3 h: 2 short courses, 13 core lessons, 1 optional.
- Use a seven-step method and a catalog of eleven patterns to design an AI solution from a blank page.
- Run a discovery workshop and turn a vague request into measurable outcomes.
- Scope an MVP with acceptance criteria, and agree what the customer's security, platform and compliance teams need.
| Module | You finish with | Lessons | Hours | Capstone |
|---|---|---|---|---|
| M11: Design method and patternsThe seven-step design method and a catalog of eleven patterns, shown on one running project. | A seven-step map and a pattern shortlist for the order and policy assistant. | 6 | 1.5 h | - |
| M7: Customer discovery and deliveryRun discovery, measurable outcomes, MVP scope, the security review, demos and rollout for the order and policy assistant. | A design pack: discovery notes, outcome sheet, scope, security checklist and rollout plan. | 7 | 1.8 h | - |
AI Solution Architect Advanced: Estimate, Propose and Defend
Estimates and total cost, build or buy, proposals and POCs, a review board, and a blank-page defense. About 2.8 h: 1 short courses, 10 core lessons, 4 optional.
- Estimate effort, timeline and three-year total cost, and compare build, buy and hybrid.
- Write a solution proposal and an executive briefing, and defend the design in an architecture review.
- Defend a timed design for an unseen brief against a scoring rubric, and plan your next 90 days.
| Module | You finish with | Lessons | Hours | Capstone |
|---|---|---|---|---|
| M10: Estimates, proposals and design reviewThe solution cycle, executive briefing, estimate and total cost, build or buy, RFP and statement of work, proof of concept and review board. | An estimate with a three-year cost table, a decision matrix, a POC plan and a review defence. | 8 | 1.8 h | - |
| M12: Blank-page defense and masteryA timed design from an unseen brief with a rubric, a 90-day practice plan and four optional drills. | A scored blank-page design and a 90-day practice plan. | 2 | 1 h | - |
How the lessons work
- Every lesson opens with one overview figure, then explains with diagrams, tables and short text.
- One running project, a fictional distributor's order and policy assistant, carries through every lesson, so each step adds to one design pack.
- Each lesson ends with an exercise, a decision-practice widget and a five-question check that explains why.
- Each lesson has an At the customer or In the interview note, so the skills carry into the role and the loop.
Start with the seven-step design method .
Which course should I take?
They are written to be used together: the Architect course teaches the design judgement, the Forward Deployed Engineer course teaches you to build and deliver, and the AI Solution Architect course teaches you to scope and propose it with a customer. Lessons link across and never repeat each other.
| Topic | AI Architect | Forward Deployed Engineer | AI Solution Architect (this course) |
|---|---|---|---|
| The question it answers | How should this AI system be designed, reviewed, operated and governed? | How do I build it in Python, ship it into a customer's environment and prove it works? | How do I turn a customer's business problem into an approved design, estimate and proposal? |
| What you produce | Requirements, decision records, diagrams, design reviews, governance controls. | A tested Python service, an evaluation report, an integration design note, and a shipped, checked deployment. | A design pack: outcomes, scope, security answers, an estimate with total cost, a proposal and a review defense. |
| How much code | Little. Mechanisms are explained, not built. | A lot. Every code lesson runs in a test suite. | None. It is about decisions, documents and conversations. |
| Format | Two courses and 41 case files. | Three levels in 10 short courses, about 17 hours, a capstone and interview question banks. | Two levels in 3 short courses, about 6 hours, one running project. |
| Start here if you are | A tech lead, architect or manager who signs off on AI designs. | An integration or backend engineer moving into customer-facing AI work. | A consultant, pre-sales engineer or engineer who has to scope and propose AI work. |
Not sure? If you already design systems and want to build them, start with the Forward Deployed Engineer course. If you already build and want the design judgement, start with the AI Architect course. If your job is to scope and propose AI work, start with the AI Solution Architect course.
In the course
- The seven-step design methodThe method every later lesson applies to the running project.
- Discovery workshopA questionnaire and a stakeholder map for the first meeting.
- Blank-page defenseA timed design on an unseen brief, scored against a rubric.
- Solution design interview questionsSizing, total cost and risk questions with a strong answer and a rubric.
- AI Solution Architect: practice quizShort checks on discovery, scope, cost and proposals.
- AI Solution Architect: practice examScenario questions with revision links for every miss.
Questions
- Who is this course for?
- Consultants, pre-sales and solution engineers, and engineers who have to scope, estimate and propose AI work for a customer. It assumes you can read a system diagram; it does not assume you write Python.
- What is the difference between Foundations and Advanced?
- Foundations covers the design method, a pattern catalog, the discovery workshop, scope and the security, platform and compliance reviews. Advanced covers estimates and total cost, build or buy, proposals and proofs of concept, a review board, and a timed blank-page design.
- Is it different from the AI Architect course?
- Yes. The AI Architect course teaches the theory: how to design, review and govern an AI system. This course is the practice of working it out with a customer: the workshops, estimates, proposals and objections. The lessons link across and do not repeat each other.
- How long does it take?
- About 6 h in total: Foundations about 3.3 h and Advanced about 2.8 h. The four design labs are optional and not counted.
- Is it free?
- Yes. The first lesson of each short course is open to everyone, and a free account unlocks the rest, tracks your progress and keeps your place.
- Are the customer, project and numbers real?
- No. The running project is a composite, labelled "Composite scenario, not a real company", and every figure in it is illustrative. Nothing in the course describes a real client or a real engagement.
- Is the compliance lesson legal advice?
- No. It explains what the common regulations and standards ask, what an engineer can supply as evidence and what only the customer's legal or compliance team decides. It never says that a product complies with anything.
- How do I know the facts are right?
- Claims about standards, products and services are checked against the primary source with the date they were checked, and the arithmetic is recomputed. The pages were drafted with AI assistance and reviewed by a second pass; none carries the author's Reviewed badge yet.