How this bootcamp works
This bootcamp is built around vibe coding: you describe what you want, and AI tools build it. Live sessions focus on briefs, decisions, and shipping. Async Deep Dives go deeper on integrations, polish, and mental models. The capstone is one product you grow every week. The rule is simple: ship one thing every week.
What building with AI actually is
The four-layer mental model (frontend, backend, data, auth) and why directing beats coding.
Tool and account setup (no code, just access)
No live-build prep required. Set up Lovable after the session, then join Maven community and WhatsApp.
Capstone seed: pick your product
Pick one small, real product with all four layers, the spine of everything you build for four weeks.
What you keep
How to go from a written brief to a working full-stack app, and how to write a brief that builds the right thing on the first try.
You ship
A live web app at a real URL, the first version of your capstone.
Watch through (async)
Think Like an AI Engineer (free 10-part series)
- 1How LLM Actually Works (in Plain English)Watch
- 2How to Prompt AI Like a Pro: The 5-Part FormulaWatch
- 3How to Pick the Right AI Model for the JobWatch
- 4Using AI With Your Files, Images and DataWatch
- 5Context Is Everything: The #1 Skill for Using AIWatch
- 6Vibe Coding for Non-Engineers (and Engineers)Watch
- 7What AI Agents Actually Are (Hype vs Reality)Watch
- 8Build AI Workflows With No Code (n8n, Zapier, Make)Watch
- 9AI Evals: How to Know If Your AI Actually WorksWatch
- 10The Fastest Way to Stay Current in AIWatch
The brief is the new spec
The durable skill of the course: write a brief specific enough that the AI builds the right thing first time.
Build your app with Lovable
Turn your brief into a running app, then iterate in the prompt-and-review loop.
Data and auth, handled inside Lovable
Add sign-up, login, and persistence by describing them, no separate backend setup.
Ship to a live URL
Publish a snapshot to a live URL, remove the Lovable watermark, then prove it works for someone who is not you.
Connecting Stripe and other integrations
Three ways Lovable connects to the outside world, app connectors, chat connectors, and any API, with Stripe as your payments example.
AI features for your app
Add live AI to your app with Lovable's built-in connector: chat, summaries, RAG, voice, and more. No API keys required.
Lovable Cloud
Full-stack hosting built in: database, auth, storage, edge functions, and AI backend. No separate Supabase setup.
Design polish for a generated app
Escape the generic AI look: design tokens in your brief, Visual Editor tweaks, external components, and visual references.
Ship your product, version one
Ship a real v1 with auth, persistence, and at least one AI feature, then share for cohort feedback and post on LinkedIn with screenshots.
What you keep
How to direct an AI coder to change a real product safely, how to read a codebase you did not write, and how to hand knowledge work to an AI collaborator.
You ship
A new feature in your product, none of it hand-written.
From "I can't code" to "I direct code"
Plan, change, review the diff, keep or roll back, the safe loop for directing code you did not write.
Extend your app with Claude Code
Run the plan-change-review loop on your capstone and confirm Week 1 still works after the upgrade.
Automate your workflow with Cowork
Hand the work around your product to an AI collaborator, same briefing discipline, new surface.
Ship a new feature
Publish the Week 2 upgrade and test it live, version one plus a directed change.
Lovable, Claude Code, or Cursor: when to use which
Start in Lovable, extend with Claude Code, reach for Cursor when you want to steer file by file.
Reading a codebase without fear
Ask for a map, trace one feature end to end, and use read-only mode to explore safely.
Ship a new feature with Claude Code
One clear change, live on your URL, directed not hand-written, and Week 1 still works.
What you keep
How to build an automation that runs without you, the real difference between a workflow and an agent, and what an agent actually is, learned by operating one.
You ship
A multi-step automation on real triggers, and a live agent you operate.
Automate with n8n
Trigger, actions, connections, build a real automation wired to your product on the canvas.
Workflow versus agent: the line that matters
Workflows follow your script; agents decide their own steps, use the simplest thing that does the job.
Operate your own agent with Hermes
Configure a persistent agent, connect it to WhatsApp, and watch it decide and act.
A peek under the hood (and where to go deeper)
A model using tools in a loop, and the honest bridge to the Engineering Bootcamp.
n8n automation patterns
Trigger-do-notify, enrich with an LLM, human in the loop, recognise the shape before you build.
Hermes skills authoring for non-coders
A skill is a described ability, write when to use it clearly, then test that it fires when it should.
The agent loop, explained without code
Think, act, observe, repeat, and why early wrong turns compound when an agent misbehaves.
Add an automation or a live agent to your product
Something useful on a real trigger, and you can say in one line whether it is a workflow or an agent.
What you keep
How to evaluate an AI feature like a leader, using TRACE. Error analysis is product work, you own the front of this loop.
You ship
Your product, evaluated, with a clear read on where it stands before the sprint.
The vibe-check trap
"It looked good when I tried it" is not evaluation, and vibes do not survive change.
TRACE: Trace and Read (error analysis is your job)
Capture real interactions, read them one by one, and journal what went wrong, product work, not engineering.
TRACE: Analyze (decide what matters, fix the obvious)
Cluster failures by frequency, fix the cheap ones, and judge pass/fail, not vague scores.
What good evals tooling looks like (so you can lead it)
Codify and Enforce are engineering, but you can recognise good checks and hold a team to them.
Build a simple must-pass checklist for your product
Turn top failures into binary pass/fail cases, and re-run the list every time you change the product.
How to brief an engineer to build the evals you need
Hand traces and must-pass cases, not a request for generic quality metrics.
Run TRACE on your product
Traces read, failures ranked, must-pass checklist built, at least one fix shipped.
Build-support office hours
Working time with help on hand. Bring a concrete blocker, not a vague "how's it looking."
Final product brief and ship checklist
One-page brief, three-minute Demo Day walkthrough, and a non-negotiable ship checklist.
Demo Day
Submit your live URL, present opt-in in three minutes, what it does, how you built it, what evals surfaced.
Resources index
- Full series: Think Like an AI Engineer (10 parts)
- Part 1: How LLM Actually Works (in Plain English)
- Part 2: How to Prompt AI Like a Pro: The 5-Part Formula
- Part 3: How to Pick the Right AI Model for the Job
- Part 4: Using AI With Your Files, Images and Data
- Part 5: Context Is Everything: The #1 Skill for Using AI
- Part 6: Vibe Coding for Non-Engineers (and Engineers)
- Part 7: What AI Agents Actually Are (Hype vs Reality)
- Part 8: Build AI Workflows With No Code (n8n, Zapier, Make)
- Part 9: AI Evals: How to Know If Your AI Actually Works
- Part 10: The Fastest Way to Stay Current in AI
- Welcome and your first project
- Prompting best practices (the brief-that-builds playbook)
- Publish your project
- Set up a custom domain
- Lovable integrations (app connectors, MCP, APIs)
- Stripe app connector
- AI features for your app
- Lovable Cloud
- Best UI with Lovable, ShadCN, React (video)
- Lovable UI polish walkthrough (video)