iBharat
connect@ibharat.tech

How we build

We build with AI, so you get software faster.

AI is part of our everyday workflow: planning, coding, testing, review and support. Our engineers stay in charge of every decision and every release.

See where AI helpsTalk to our team

Our workflow

AI at every step, people at every decision.

This is how a feature moves from idea to production in our team.

  1. Plan with AI

    We write the requirements and a step-by-step build plan with AI, then review and agree on them with you before coding.

  2. Build with AI

    Engineers use Claude Code and GitHub Copilot to write code, following each project’s written rules for AI assistants.

  3. Test as we go

    Tests are written alongside each feature and run automatically with every change.

  4. Human review

    Every change is reviewed by an engineer before it is merged.

  5. Automated releases

    GitHub Actions builds Docker images for each release, deploys them to a test environment first, then promotes the release to production.

  6. Support with AI

    After launch, AI helps us read logs and find the cause of problems quickly. Every fix is reviewed by an engineer before it ships.

AI tools we use

  • Claude Code
  • GitHub Copilot

Where AI helps

Four places AI saves us time.

This isn’t a product we sell. It’s how we work on our projects every day.

  1. 01

    Planning

    Turning a brief into clear requirements, edge cases and a build plan faster.

    • Requirements
    • Build plans
    • Edge cases
  2. 02

    Writing code

    Setting up screens, APIs and database changes quickly, then improving the code safely as the product grows.

    • Claude Code
    • GitHub Copilot
    • Project rules
  3. 03

    Testing & review

    Drafting unit and browser tests and flagging risky changes before an engineer reviews them.

    • Unit tests
    • Playwright
    • Code review
  4. 04

    Support & Maintenance

    Reading logs, tracing bugs, upgrading dependencies and keeping documentation up to date.

    • Logs
    • Runbooks
    • Upgrades

Our stack

Technologies chosen for each project.

These are the tools we build with. Each project uses only the ones it needs.

How a typical iBharat product is built

  • Web app

  • Mobile apps

API

  • Database

    Storage bucket

  • Messaging & video

  • Payments

Runs on AWS

CI/CD pipeline

  • Web

    The screens people use in a browser: websites, customer portals and admin dashboards.

    • Next.js
    • React
    • TypeScript
    • Tailwind

  • Mobile

    Android and iOS apps written once in a cross-platform framework and published to both stores.

    • Flutter
    • React Native

  • Backend

    The server side: APIs the apps call, business rules and scheduled jobs such as reminders. Odoo where a client already runs on it.

    • NestJS
    • Node.js
    • Python
    • Odoo

  • Data

    Where records are stored: PostgreSQL or MongoDB for the main data, Redis for fast, short-lived data such as caches.

    • MongoDB
    • PostgreSQL
    • Redis

  • Cloud & DevOps

    Docker packages each app, GitHub Actions builds and deploys it, and AWS servers with nginx run it.

    • Docker
    • GitHub Actions
    • AWS
    • nginx

  • Messaging & video

    WhatsApp and SMS messages, push notifications on phones, and in-app video calls.

    • WhatsApp API
    • Firebase
    • LiveKit

  • Payments

    Online payments for bookings and lab tests.

    • Razorpay

Have an idea? Let’s build it fast.

Tell us what you need. We will show you how we would build it and how long it would take.

Start your project