Gohan AIGohan AIGet early access

01/AI for fitness · Buenos Aires


An AI personal trainer your gym’s app can plug in. Conversational coaching, routines that rewrite themselves in real-time, integrable in any existing app via MCP.

Book a 15-min walkthrough
Try the live demo

02/Built for the apps your members already open

Megatlon
SportClub
Always
Smart Fit
Fiter
Anytime Fitness
Crunch
Megatlon
SportClub
Always
Smart Fit
Fiter
Anytime Fitness
Crunch

03/Manifesto

Gym apps haven’t kept up.
Routines stay frozen. Edits feel like
filing a tax form. Members close the
app and open ChatGPT instead.

We close that gap. Gohan AI ships your members a coach that knows their level, their equipment, their injuries — and rewrites their programme as they live their week.

No new app. No new login. Inside the brand they trust. Your backend, your data, your members. Our intelligence.

04/Three pillars

I.

Coaching that
converses.

Streaming AI replies, voice or text. The only personal trainer that answers at 2am when your member is doubting their squat depth.

II.

Routines that
rewrite themselves.

Members ask. The plan changes. The app updates live. Conversation maps to structured database mutations through Claude tool use.

III.

Drops into
any app.

We ship as a module + MCP server. Your engineers wire it into your existing app in an afternoon. Your brand. Your members. Our brain.

05/Try it · in the page

Scroll. The app
opens in front of you.

Below is the real product, not a video. The demo account is already signed in — tap inside, chat with the coach, watch the routine rewrite itself in real-time.

Scroll to open the app

06/How it works

Three steps.
One afternoon.

01

Connect

Embed our React module in your app, or point your backend agent at our MCP server over HTTP. Both paths use the same nine-tool API.

02

Configure

Issue a tenant API key. Pass logo, colors, and your user mapping. Tenant scoping is automatic from there — your members never leave your brand.

03

Coach

Members chat naturally. Our AI calls structured tools to mutate plans. Realtime pushes the changes back to every screen.

07/For developers

Two ways in.
One brain.
Your data, untouched.

Drop the React module into your gym app for an embedded chat, or point your own backend agent at our MCP server over HTTP — both paths land on the same nine-tool API. Your members keep your brand, your data stays in your Supabase.

  • ▸chat module · React + Gohan session JWT
  • ▸MCP server · HTTP + tenant API key
  • ▸9 tools across both paths
  • ▸tenant-scoped, automatic
  • ▸realtime via Supabase to every screen
MemberApp.tsx
// Drop Gohan AI into your gym app
import { GohanCoach } from "@gohan-ai/react";

export function MemberApp({ user, brand }) {
  return (
    <GohanCoach
      memberId={user.id}
      tenant={{
        slug: "your-gym",
        primaryColor: brand.primary,
        logoUrl: brand.logo,
      }}
      // Member chats. Plan rewrites
      // itself. Realtime to your DB.
    />
  );
}

08/Numbers we obsess over

<1d
Time to integrate the SDK
drop-in component
0
Tools the AI calls to mutate plans
multi-step loop
0ms
From AI edit to routine on screen
supabase realtime
∞
Tenants per deployment
RLS-scoped, multi-brand

09/Talk to us

Your members are
already chatting
with an AI.
Make it yours.

Start the conversation
See the code