We build software, then we run it.

A software company in Kochi, India. We build our own mobile apps, web platforms, APIs and the cloud under them.

Contact Now building

Since 2022 Registered in Kerala, India

One team, end to end Design, build, deploy, support

We run the servers AWS, Google Cloud and Firebase

Now building A gold savings app

Now building

Our own product. The same team designs it, builds it and runs it.

A mobile app for monthly gold savings in India. In development.

The parts our software is made of.

Sign-up, identity checks, payments, notifications and a profile. Built for a slow phone on a slow network, because that is what many users have.

The screens staff live in: work queues, approvals, search, bulk actions, exports and an audit trail that says who changed what.

Collection, reconciliation against the bank, invoices, refunds and the ledger that has to add up at the end of the month.

Daily numbers staff trust, scheduled jobs that build them overnight, and exports in the format the accountant actually wants.

Payment providers, bank statement files, messaging, identity services, CRM and any third-party API with documentation and a sandbox.

Reading receipts and documents with vision models, checking uploads against our records, and classifying what comes in.

Five parts. Working software from the second week, so the launch date never comes as a surprise.

Discover

We ask what the software must do, who uses it and what breaks today. The answers go into a written scope.
Design

Screens and data model first. We click through a prototype before anyone writes the expensive part.
Build

Weekly releases to a test environment. Written tests on the parts that handle money.
Launch

Store review, cloud setup, data migration and a rehearsal. We launch on a working day, never on a Friday evening.
Operate

We keep running it after launch: releases, backups, monitoring and cost.

How we build

Coding agents work on everything we build. That is how a small team ships like a larger one.

A faster first draft only pays if the review stays as careful as before.

Agents work inside written project rules: the glossary, the architecture notes and the coding standard. A human reads every change before it merges.

Tests, review and the deploy gate stay exactly where they were. Agents never release to production on their own: a person approves every deploy.

We write the skills, the plugins and the pipelines our agents run on. So we know where they help, and we know where they are confidently wrong.

We keep a small, settled stack and know it well. Fewer surprises in year three.

Languages: TypeScript, Python, Go, SQL
Mobile: React Native, Expo, Google Play, Firebase
Web: Next.js, React, Tailwind, Vercel
Backend: Django, NestJS, Go services, GraphQL, REST, background workers
Data: MySQL, PostgreSQL, Redis, scheduled pipelines, reporting
Cloud: AWS: ECS Fargate, RDS, S3, CloudWatch, Route 53, Secrets Manager. Google Cloud and Firebase. Docker, GitHub Actions.
AI: Claude and Claude Code, vision models for document checks, our own agent tooling