
Founder @ JS Mastery @GitHub Star, software engineer & educator Agentic Engineering Course is live now, 20% off through Sep 29
Croatia
Joined January 2019
- Tweets6K
- Following181
- Followers42.3K
- Likes3.9K
Pinned Tweet
NEW VIDEO: Build a Full Stack Airbnb Clone with React Native and AI Agents 🚀
➡ youtu.be/GXbZc1sHEJY
“Quiet cabin in the woods, hot tub, dog-friendly, under $300 for two.”
Type that. The app finds it. No filters.
Not even Airbnb has this.
Here’s what you’ll build 🧵👇
OpenCode got hacked this weekend, and they called it a “distillation attack.”
Here’s what that means 👇
Someone asks a big AI model like Claude millions of questions, saves every answer, and trains a cheap copycat model on them.
AI labs block it, so attackers steal real developers’ API keys instead. They get the answers. You get the bill.
Some OpenCode users got hit for $1,700.
We had a security breach last night related to a large scale distillation attack that impacted 1.4% of OpenCode Console users
We have notified them all this morning via email. If you did not get an email you were not impacted, it was a subset of users who had external providers linked
The incident is ongoing so we cannot share more details at this point but we are working with Anthropic, OpenAI, and Google to remediate
Long before AI agents could write code, serious software projects already followed a familiar path.
You work out what you're building. You make the important technical decisions and write them down.
You set the conventions everyone follows. You build one feature at a time. And before anything gets called finished, you prove it works.
That's the software development lifecycle.
Requirements into design > design into implementation > implementation into verification.
Every stage producing something the next stage needs.
It is older than Claude, older than Cursor, older than any AI coding tool.
Even writing decisions down isn't new.
Engineers have kept architecture decision records for years, for one very practical reason. If the reasoning behind an important choice only lives in somebody's head, it eventually disappears.
Careful teams don't rely on memory. They leave evidence.
The JSM workflow skills don't invent a new way to build software. They make the AI coding agent follow the process a good engineering team already follows.
One stage at a time, each decision feeding the next.
Grab the skills for free here ➡ jsmastery.com/skills
We built the same app twice.
Once vibe-coded then Once with a proper agentic workflow.
Year-one cost: ~$43K vs ~$3.5K 🤯
“It works” and “it’s cheap to keep alive” are very different things.
Full breakdown in our Agentic Engineering course 👇
jsmastery.com/course/agentic…
Ever noticed your AI coding agent getting dumber after an hour?
That’s context rot. The longer the session, the more it forgets what actually matters.
Fix: smaller tasks, fresh sessions, clear specs.
We cover how agents really work (free lesson included) 👇
jsmastery.com/course/agentic…
“Built it in one prompt.” - Cool. How many iterations did it take to make it actually good?
AI gives you the first draft. Your judgment makes it a product.
How many prompts did your last feature take? 📷
What if you could find the perfect stay just by describing it? 📷
"Cozy cabin near a lake, under $150 where I can code in peace" → real results.
Build Roam, an Airbnb-style app, with AI agents + React Native, Clerk, Sanity & PostHog.
Full tutorial:
youtube.com/watch?v=GXbZc1sH…
Does careful engineering still matter when AI writes the code? Everyone has an opinion. Almost nobody has data.
So I ran a controlled experiment: I built the exact same app twice in one day using two completely different workflows.
Here’s what happened:
The Project: A Movie Recommendation Engine
1️⃣ Build 1: Pure Vibe Coding
• Prompt-as-you-go
• No written plans or tests
• Allowed the AI agent to improvise architecture
2️⃣ Build 2: Structured Engineering with the JSM Workflow
• Scope captured first
• AI gated from unauthorized architectural choices
• Full test coverage, migrations, and disciplined setup
The wild part? The vibe build was impressive.
If I demoed it cold, you'd love it. It worked.
But demos aren't production.
What happens at 5,000 users? Or when you open the codebase 6 weeks later to refactor?
Are we actually accelerating development, or are we just shipping future technical debt at record speed? 🤔
One thing I’ve started liking about @posthog: I don’t always have to build every chart myself.
If I already know what I want to learn, I can just ask for it in plain English. 🧵👇
4/ I can also ask it to build a whole signup dashboard.
Or just ask what’s interesting in my app right now and see what it finds.
5/ It saves me a lot of clicking.
Watch @posthog in action in our latest YouTube video ➡ youtu.be/GXbZc1sHEJY
5/ The stack:
@reactnative + @expo + @typescript
@nativewindcss
@clerk for auth
@sanity_io for content, embeddings and knowledge bases
@posthog for analytics
@claudeai
6/ And - the app can change.
✅ Startup
✅ Client project
✅ Marketplace
✅ SaaS.
But the process stays the same.
Watch the full course free on YouTube ➡ youtu.be/GXbZc1sHEJY
You are the architect. The agent is the engine.
You stop being the person hoping for good output and become the person who decides what good output even is.
You set the constraints, you make the real decisions, you say "no, not like that, like this," and you understand every choice that gets made.
The AI is the engine: astonishingly fast, tireless, more code in an hour than you could type in a day.
But it does not have judgment.
Judgment is your job.
It always was.




