Two days. From individual chaos to a shared method.
A hands-on intensive that takes your engineering team from ad-hoc AI use to a structured, governable practice. Day 1: build trust. Day 2: build capability.
The role of the developer has fundamentally changed.
GitHub Copilot now writes 46% of code for 15 million developers. AI handles the repetitive work. Your developers need to focus on what AI can't do: architecture, design decisions, quality control.
Industry leaders agree: "Developers will become architects of distributed orchestrated intelligence." Like a conductor who doesn't play every instrument but creates the symphony — your developers become more valuable, not less.
Vibe coding creates "the tech debt of fifty engineers." AI-generated code has 41% higher churn and 48% contains vulnerabilities. Your team needs a production-grade system — consistent, repeatable, architecturally sound.
Companies with AI-fluent teams deploy 3x faster with 50% fewer failures. The teams winning aren't replacing developers — they're elevating them. Architects ship 2-5x faster with AI as their builder.
The CRAFT Workshop: where AI-native development becomes real.
CRAFT Method is the first method purpose-built for AI-native development — 5 active practices (Curate, Refine, Architect, Fortify, Test). This workshop is where your team experiences them hands-on, with AI as the execution layer and your developers as the architects.
"Gartner predicts 80% of developers will need AI skills by 2027. They're wrong. You need them now."
Companies that invest in structured AI training see a 26% productivity boost vs chaotic adoption that often results in 19% slower delivery.
Building a platform alone taught me what developers really need.
I built a complete software platform — 11 integrated products, hundreds of thousands of lines of code — with AI as my only team member.
I was a developer years ago. I stopped coding day-to-day. But when I came back to build this platform, I realized something: the skills I'd kept — systems thinking, architecture, orchestration — were exactly what AI needed from me.
The coding itself? AI handled that. What AI couldn't do was decide what to build, how to structure it, and when to override its suggestions.
That experience gave me a realization: developers don't need to be replaced — they need to evolve.
The best developers will become architects who direct AI. They'll 10x their productivity, not lose their jobs. But they need to learn a new skill set. This workshop exists because I want to help organizations make this transition. Your developers have the foundation. They just need to learn how to conduct.
— Luis Gonçalves · Founder, CRAFT Method
One architect with AI ships more than ten coders without it.
This isn't about replacing your team. It's about elevating them to where they can deliver 10x more value.
2 days to practise the 5 CRAFT verbs.
Morning is theory mapped to Curate, Refine, Architect, Fortify, Test. Afternoon is practice. Every developer ships real code both days.
- Why "typing code" is no longer the job — the architect mindset
- Curate in practice: AGENTS.md / CLAUDE.md, codebase intelligence, context governance
- Refine in practice: spec completeness, predicate-form criteria, AI edge cases
- AI capabilities and limitations — what to delegate, what to keep
- Claude Code setup, slash commands, dev-workflow integration
- Fix real bugs in open-source projects with AI
- Curate live: write a CLAUDE.md / AGENTS.md, layer context, score quality
- Refine live: write a complete predicate-form spec for each fix before AI generates code
- Iterate, validate, see quick wins that build trust in the verbs
- Everyone ships working fixes by end of day
- Architect in practice: pattern libraries, ADRs, multi-agent code review, agentic CI/CD, system topology
- PRP = PRD + Codebase Intelligence + Validation — prompts that succeed first pass
- Breaking features into AI-executable chunks; how to create & manage agents
- Fortify in practice: SAST/DAST/SCA at AI velocity, SBOMs/AIBOMs, supply-chain defense, ISO 42001 / EU AI Act readiness
- Test in practice: AI test generation, mutation testing, edge cases, quality gates that block bad merges
- Ship real features with AI (your choice: sample projects or your codebase)
- Architect live: enforce patterns, ADRs, and multi-agent review on every PR
- Fortify live: run SAST/SCA at AI velocity, generate SBOM/AIBOM, remediate findings
- Test live: AI-generated tests, mutation runs, edge-case probing, quality-gate merge
- Code review session and your team playbook for Monday
The 5 practices of AI-native development.
Prompt engineering is now a "core developer skill." These are the five active practices that define AI-native development.
- Authoring AGENTS.md / CLAUDE.md for your stack
- Codebase intelligence + context governance
- Dev-workflow and onboarding context for the team
- Predicate-form acceptance criteria before any code
- Naming AI edge cases up front, not after a bug
- Spec-first workflow with traceability to code
- Pattern libraries and ADRs as guardrails for AI
- Multi-agent code review and agentic CI/CD
- System topology and architecture evolution
- SAST/DAST/SCA at AI velocity, vulnerability remediation
- SBOMs/AIBOMs and supply-chain defense
- ISO 42001 and EU AI Act readiness
- AI-generated tests, mutation testing, edge-case probing
- Quality gates that block bad merges
- Deployment quality and feedback loops back into Curate / Refine
Everything your team gets.
- 2-day intensive (in-person or virtual)
- Pre-workshop discovery call
- Customized to your stack and codebase
- Hands-on bug fixing exercises (Day 1)
- Hands-on feature building exercises (Day 2)
- Live coaching and Q&A throughout
- 30-day email support
- Implementation guidance
- Week 3 check-in call
- Week 4 review session
What CTOs ask.
Can this be delivered remotely?
Yes. In-person or virtual. Same impact, same exercises, same results.
What's the ideal team size?
3–20 developers per session. Larger teams? We run multiple cohorts.
Do we use our actual codebase?
Day 1 uses public open-source projects (builds confidence with low risk). Day 2 can use your codebase if you choose — that's when trust is established.
What if our organization restricts AI tools?
We work within your security policies. Many enterprises use Claude with data retention controls.
What about the investment?
Contact us for a custom quote based on team size and location. This is a transformation, not a course — priced accordingly.
For leaders ready to make AI-native development an organisational capability.
- Lead a team of 3–50+ developers ready to level up
- Want AI-native development as a team-wide capability
- Believe AI should amplify developers, not replace them
- Need a consistent method across your team
- Want hands-on training, not passive video watching
- Are ready to invest in your team's evolution
- You see AI as a way to reduce headcount
- You expect transformation without practice
- Your organization bans AI tools entirely
- You're not ready for a method — you just want quick tips
- You're looking for a silver bullet
Start your team's AI-native development journey.
Your team already writes code. CRAFT Method gives them 5 active practices to build with AI — properly. Two days. Hands-on practice. Real code shipped. Your organisation leaves with a structured capability, not just individual skills.