Pro Vibe Coding Setup: stop re-explaining, stop hitting limits.

Bring at least one Pro-level plan for the Claude Code or Codex parts: a paid Claude plan (Pro or above), or a ChatGPT plan that includes Codex (Plus or above). Antigravity has a free tier, so that part costs nothing to start. The ₹199 covers the session only and does not include any plan. Plan contents and prices change, so check each vendor's page.

No. Most people use one, and that is fine. The session works through each tool separately, so you can focus on the one you use. The part about sharing one memory across tools only matters if you run two or more.

Two hours will not make you an expert in all six, and we do not pretend it will. Each block is hands-on and short: you leave with a working setup and templates for the tool you use. The recording, checklist and kits carry the depth, and the 4-week cohort goes further for people who want it.

Videos show you someone else's screen. This is a live session on your own laptop, where install errors and the rules file get fixed with you in the room. The small fee also keeps the room full of people who actually turn up and build.

There is a 4-week cohort, and we will tell you about it in the last ten minutes — once, with the details and an application. It is optional and selective, and nothing earlier in the session depends on it.

If you already use an AI coding tool, yes. You direct the agent in plain language; the only typing is prompts and a few terminal commands we run together, slowly, with help when something breaks. If you have never used one, try a free tier first, because this session builds on that.

Antigravity is Google's agent-first coding tool, and free personal accounts get a weekly quota. Eligible Jio users were offered 18 months of Google AI Pro, which raises its limits, but that offer's claim window appears to have ended in late October 2026, so check the MyJio app. If you hold it, we set Antigravity up with it; if not, we use the free tier. Which models the free tier includes keeps changing, so we show you how to check.

Many AI coding plans meter your usage in rolling windows, commonly around five hours, after which your allowance resets, and some add weekly caps. Run out mid-task and your work stalls. Exact limits differ by plan and change over time, so we do not teach numbers to memorise: we teach how to check where you stand and how to spend a window on the work that earns it.

Your tool forgets everything between sessions unless it is written down. Central memory is one organised place for your project rules, decisions and handoff notes that every session reads at the start, so the agent begins informed instead of blank. We set one up on your project and cover how to keep it short enough to stay useful.

A code graph is a queryable map of your codebase: its files, functions and how they call each other. With one, an agent can look up what it needs instead of reading whole files again and again, which spends fewer tokens and so burns less of your window. It helps most on larger projects; on a small one a good rules file may be enough, and we say how to tell. We build one on your project in the session.

Most wasted tokens come from re-sending context the tool could look up, vague asks that need several rounds, and long chats that keep re-billing old messages. We practise the fixes on your own prompts: point at files instead of pasting them, state the goal and the done-when up front, ask for a plan before code and for diffs instead of full rewrites, and start a fresh session on purpose. Savings vary by project, so we show how to measure yours instead of promising a number.

Yes, usually: two agents working at once spend a window faster than one. Overlapping pays off when the tasks are independent and you would otherwise be waiting, and not when it just makes you hit the limit sooner. We cover how to tell the difference, and how to keep parallel work on separate branches or worktrees so it does not collide.

One tool's chat history does not travel to another, so the only memory that crosses tools is what is written down in your repo. We set up one rules file the tools share, plus a short handoff note whenever work moves between them, so the second tool starts where the first stopped.

The recording, checklist and templates arrive within 24 hours. Live is where you get unstuck in minutes, so come if you can.

Full refund before the session starts, no questions. See the refund policy.