Why Growtricity — an in-person AI course in Chandigarh where you build real software by directing AI

Why Growtricity

Yeh koi jaadu ki pudiya nahi.

This is no shortcut. These are the same techniques 100+ students across Mexico and the USA earnestly learned and put to work — to grow their careers, their businesses, their communities. For the first time, that comes to Chandigarh: so you don't just catch up with the world — you lead it, with the techniques the top 1% of builders use.

How you'll actually learn

You direct the AI. You understand every line.

This is the real skill — the world's top AI people call it agentic engineering, and it's where software is heading. Not "the AI writes your code and you hope it works." You direct AI to build, then read it, explain it, and fix it yourself — the part employers actually look for, and the part almost nobody here teaches.

"You can outsource your thinking, but never your understanding."Andrej Karpathy · OpenAI founding member · AI Ascent 2026

Direct the AI

Tell it what to build, give it the right background, pick the right model for the job. The discipline behind it — not just which buttons to press.

Understand what you ship

The AI writes the code — this is engineering, not hand-coding. But every class you explain what you built: the system, the decisions, why it works, how you'd change it. Can't explain it yet? You don't move on. Understanding what you ship — not blindly shipping it — is what makes you employable.

Ship something real

Every level ends with a real product on a real web address — something a client or an employer can open and judge in a minute.

What you'll learn

The toolkit real builders use — explained from scratch.

Across the three levels, in plain language and from the ground up. You build a real, working understanding of how this all works — not a surface tour. The hard parts — the ones that actually matter in real work — are the whole point.

Build AI you can trust

✓Stop the AI from making things up — using a method the instructor's team built and proved in real businesses in Mexico (developed with lawyers, where one invented fact has real consequences).
✓Making AI reliable enough to put in front of real customers is the hard part — and the whole point. This is the part a chatbot tour never reaches.

The AI toolkit — and the judgment to use it

✓Which AI model to trust for which job — open vs closed models, which are free vs paid, and the honest tradeoffs in cost, speed and quality.
✓When a free building block is enough, and when a paid one is worth it.
✓How to direct AI like a professional — not just chat with one bot.
✓When to give the AI reusable "skills" — saved instructions that make it produce better, more consistent work — and which ones are worth building.

Work like a real engineering team

✓Track every change cleanly and write docs anyone can follow — the standard way teams work.
✓Keep "technical debt" — the shortcuts that slow a project down later — under control.
✓Plan like pros (Kanban boards, Agile vs Waterfall) and test properly before users see it.
✓Manage your time the way real builders do — how much goes to coding, learning, research, brainstorming, and working problems out with other people.
✓The working habits used at Google, Apple, Meta, Microsoft, Tesla and Airbnb.

Understand how AI works — and build agents

✓Open-source vs closed-source AI models — what each really is, the honest differences, and when to use which.
✓How AI models actually learn, in plain language — neural networks, weights and fine-tuning, without a science degree.
✓Build software agents that do real work on their own — and shape how one behaves for a specific job.
✓Enough real understanding to keep up as AI keeps changing — and adapt what you build.

Choose the stack, run the infrastructure

✓Choose the right tools and database for the job — by reasoning, not copying a tutorial.
✓Put your software online and keep it running safely — so it's not just on your laptop.
✓Keep it simple enough to manage, solid enough to do the job.

Design for real people, then ship

✓Read what a business needs, and find the quick wins first.
✓Design software people use without thinking — clear, low-effort, genuinely good UX.
✓Know when it's truly ready for users, release it, and keep it working as more people come.

Tools you'll use: Cursor Claude Code Codex GitHub the command line Linux cloud servers Convex (app database) Clerk (logins) Tailscale (private networking) Resend (sending email) Add Google/Facebook sign-in Google Sheets Google Drive

The difference

What makes Growtricity different.

Not louder promises — a different way of building, and proof you can check.

How you learn

✓You build real software by directing AI — and leave with a link you can open.
✓A small room of 12, in person, taught by a founder who has shipped and sold real software.
✓The 2026 way the world's best builders actually work.

Why you can trust it

✓A real, checkable track record — 100+ students taught across the US and Mexico.
✓Live software you can open right now, and a registered company you can look up.
✓Meet the instructor and see the work before you decide.

The course · 3 levels, about 3 months

From "I know a little code" to "I ship real software."

Three days a week, about five hours each — roughly 190 hours of building, all in person. ₹15,000 for all three levels.

Level 1 · Month 1

Build & ship

Go from "I know a little code" to two real products, live on the internet.

The developer toolkit college skipped — the terminal, git, Linux, SSH
How to actually work with AI: which model for which job, giving it the right context, catching its mistakes
The tools the pros use — Cursor, Claude Code, Copilot — and how to drive them
Build a clean, real website, live on its own web address
Build a full app with a login and a database, also live

You leave with: two live products + a GitHub anyone can open.

Level 2 · Month 2

Real software

Build the kind of software a business actually pays for.

Logins and accounts done properly — the same setup the instructor's live apps use
Keeping every customer's data separate and safe — you build it, then try to break it
Roles, permissions, and billing — free vs paid plans
Add features to safe copies of the instructor's real apps (Peter Diary, Peter CRM)
Launch a production-grade app in the real world — live, on a real domain

You leave with: your own real software product a business could pay for.

Level 3 · Month 3

Automate & launch

Turn the skill into real, shippable products — and become job- and client-ready.

As a class, take a brand-new product all the way to live — research what people actually want, choose the right tools and libraries (the tech stack), and build the command center, the one dashboard it all runs on
Automation: software that runs on its own, on a schedule, no one watching
AI that answers from your own documents — not made-up answers
A working WhatsApp agent, live on a real number
Present your work: scoping and quoting real projects, proposals, handling clients
Your portfolio packaged + applications out — to jobs and clients alike
A special guest class with a working industry expert — from India or abroad.

You leave with: real software you've built and put online, a portfolio to show any employer or client, and you've already started applying — for jobs and freelance work.

Ashish Punj, Founder and Lead Instructor, smiling beside a Growtricity billboard — Ashish Punj · Founder & Lead Instructor · Chandigarh

Who teaches

Taught by Ashish Punj — the international instructor who built the app you just opened.

Ashish Punj. 30 years building and selling software, working with companies across the US and Mexico. He built Pikkop, a logistics platform in Mexico that moved over a million packages, and sold it in 2024. This isn't an institute with 200 names on a Zoom call — it's a small room of 12, and he teaches every class himself.

Schooled in Chandigarh, he's taught AI to lawyers, founders and operators across two continents — now he's teaching it here, to people who want to build.

He's already taught 100+ students across Mexico and the USA — people who took these same techniques and grew their careers, businesses, and communities. Now it's Tricity's turn.

30 years · US & Mexico Built & sold Pikkop Wharton · Operations OpenAI · ChatGPT for Teachers Anthropic · Claude AI Fluency Google · Business Intelligence IBM · Generative AI EN · हिंदी · ਪੰਜਾਬੀ