01 / 13
adaptto.ai · claude code workshop

From AI visitor to agentic operator.

A practical training workshop for teams who don't have months to figure this out.

01 — the problem

Staying current with AI is a full-time job.

New tools, new techniques, new best practices every few weeks. Most teams either fall behind or burn hours chasing every update themselves — and neither is a real strategy.

02 — the solution

Skip the months of trial and error.

Compress
Months into 4 weeks
The mistakes, the dead ends, the "wish someone told me that" moments — already worked out.
Become
An agentic operator
Not just a better prompter — someone who ships real work with Claude Code.
Create
Real leverage
Work that used to take a day, done in an hour — by someone already on your team.

Often pays for itself inside the 4 weeks. By week 3, it's common to see participants building things that used to require bringing in outside help.

03 — what it is

4 weeks. 4 live lessons. One platform.

Format
4 live lessons + async work in between. ≤2.5h/week per participant, ≤1.5h live.
The platform
Each session lives on the AdapttoAI Academy platform — read it in chunks, answer a quick check per chunk, track progress on a stepper.
Audience
Non-technical operators — no engineering background required.
Live work
Every live lesson includes hands-on building on a real project, not just watching a demo.
04 — the curriculum

The arc: plan → build → extend → master.

L1
Plan
The planning mindset. Leave with a high-quality, context-rich plan — execution starts next.
L2
Build
Files are Claude's backbone. Execute the plan in phases, on a real project.
L3
Skills, Rules & MCPs
The global layer, workflow-level skills, and when to wire an MCP.
L4
Agentic capstone
Agent-human parity, verification before "done," and a capstone demo.

One real project threads through all 4 weeks — participants build it start to finish, not four disconnected exercises.

05 — the platform

Track your pre-work before every class.

Interactive readings and quick checks before each session. Live time goes straight to building. Try the quiz →

academy.adaptto.ai
Claude Code Workshop · Your Cohort
✓Plan
2Build
3Skills
4Agentic
Lesson 2 · Build on the Plan
Quick check — what does Claude re-read every prompt?
CLAUDE.md
README
I'm done ✓
06 — built by

Who teaches it.

Co-Founder
Giuseppe Belpiede
  • Founder of LATAM's biggest hiring marketplace for developers, sold to Series C US company (Revelo).
  • Managing Director at various ventures globally at Rocket Internet and Revelo, with a growth focus (Marketing / Sales / M&As).
  • 15+ years in technology, fully dedicated to learning and teaching AI since end of 2025.
LinkedIn →
Co-Founder
Raffaello Starace
  • Founded and scaled an e-commerce company to $100M+ USD (and raised $100M+ in VC funding).
  • Consultant at Booz & Co (now PwC).
  • 15+ years running operations, mainly in tech companies, now building the tools to automate them.
LinkedIn →
07 — the cadence

How a week works.

01
Live session
≤1.5h, 10–15 min Q&A to close.
02
Async prep
Set up the next lesson before it starts.
03
Recording
Delivered within 48h post-session.
04
Cohort channel
WhatsApp or Slack — questions answered within 1 business day.
08 — the throughline

What participants build.

L1
Plan it
A real working example, planned in detail.
Pre-L2
Instantiate it
Project set up, first CLAUDE.md drafted.
L2–L3
Build & extend
Execute in phases, add skills, rules, MCPs.
L4
Capstone
Each participant demos what they shipped.

Same project, start to finish — not four disconnected exercises.

09 — the shift

What changes for each participant.

01
Plan-first mindsetBefore building anything — most of the value lives in the plan.
02
Files-as-memory modelCLAUDE.md, docs, and skills are how Claude remembers — improving the files improves Claude.
03
Reusable skillsRepeated gestures become saved, composable workflows.
04
Verification judgmentKnowing when to trust the output, and when to check it.
10 — in short

What they walk away with.

Setup
A working Claude Code setup
Personal, configured, ready to keep using after the cohort ends.
Artifact
One shipped asset
Real, not a toy exercise — built and demoed at the capstone.
Mindset
Compounding mental models
The judgment to keep improving without us in the room.
11 — why this exists

“We work primarily in the Claude ecosystem, and we'll teach your team to do the same. The mental models transfer to any AI suite they encounter after. This is what we found, packaged so your team doesn't start from zero.”

next step

Want to scope a cohort for your team?

Pricing discussed on the call.

Book a call →
1.
30-min scoping call.
2.
We tailor the plan to your team.
3.
Kickoff.