Technology
Our technology stack, selected for long-term maintainability
Every choice is tested against four criteria: long-term vendor and community support, depth of the hiring market, fit for the specific problem, and cost to operate. The answer is usually a well-established tool.
Technologies we work with
Grouped by layer. Each entry explains when we select it and when we do not.
Frontend
React
Our default for interactive interfaces: huge ecosystem, easy to hire for, and stable enough to bet a product on.
Next.js
React with server rendering, routing and image handling built in. Fast public sites and web apps from one codebase.
TypeScript
JavaScript with types. Catches a whole class of bugs before they ship and makes large codebases easier to change.
Tailwind CSS
Utility-first styling that keeps design consistent and stylesheets small as a product grows.
Mobile
React Native
One codebase for Android and iOS, sharing skills and sometimes code with a React web app.
Flutter
Cross-platform apps with consistent rendering on every device, well suited to design-heavy interfaces.
Kotlin (Android)
Native Android for apps that lean on hardware, background work or Bluetooth.
Swift (iOS)
Native iOS when an app needs platform features cross-platform tools handle poorly.
Backend
Node.js
JavaScript on the server. Good for APIs, real-time features and sharing types with the front end.
NestJS
A structured Node.js framework that keeps larger back ends organised into clear modules.
Python
Our choice for data processing, AI pipelines and scripts that glue systems together.
FastAPI
A quick, well-typed Python framework for APIs, especially those serving AI or data features.
PHP and Laravel
Mature and widely hosted. We maintain and extend Laravel systems and build new ones where it fits the team.
Databases
MySQL
Reliable relational database, available on almost every hosting plan, and the default for many business systems.
PostgreSQL
Relational database with strong support for JSON, geospatial data and vector search.
MongoDB
Document database for flexible data shapes. We use it when the data genuinely lacks a fixed structure.
Redis
In-memory store for caching, queues and rate limiting when a system needs to respond quickly under load.
AI & Data
Anthropic Claude API
Large language models for reading long documents, drafting and careful reasoning tasks.
OpenAI API
GPT models for chat, extraction and embeddings, with a broad set of tools around them.
Google Gemini API
Multimodal models that work with text, images and documents, integrated with Google Cloud.
Vector search (pgvector, Qdrant)
Finds passages by meaning rather than keywords, the backbone of assistants that answer from your documents.
pandas and the Python data stack
Cleaning, reshaping and analysing business data before it feeds reports, forecasts or AI features.
Cloud & DevOps
Amazon Web Services
Broad cloud platform with Indian regions, used when a system needs managed databases, queues or storage at scale.
Google Cloud
Cloud hosting with strong data and AI services, also with Indian regions.
Docker
Packages an application with its dependencies so it runs the same on a laptop, a test server and production.
GitHub Actions
Automated tests and deployments on every change, so releases are routine rather than nerve-racking.
Cloudflare
DNS, caching, bot protection and spam-resistant forms in front of websites and apps.
IoT
MQTT
A lightweight messaging protocol built for devices on unreliable networks.
ESP32
Low-cost microcontroller with Wi-Fi and Bluetooth, behind many sensor and control devices.
Raspberry Pi
Small Linux computer we use as a local gateway, kiosk controller or edge processor.
Other
WordPress
Still the right answer for some content sites, especially when a team already knows how to use it.
Shopify
Hosted e-commerce that lets a brand start selling quickly, extended with apps and custom integrations.
WhatsApp Business Platform
The official API for automated WhatsApp messages, notifications and assistants.
Razorpay
Indian payment gateway for UPI, cards, net banking, payment links and subscriptions.
Selection criteria
Maintainability after handover. If you intend to build an in-house team, we choose technology with a deep hiring pool in India. A niche framework is a long-term staffing risk.
Minimal moving parts. Every additional service must be patched, monitored and paid for. We add components only when the requirement justifies them.
Open standards and portability. Standard databases, documented APIs and containerised deployment allow you to change hosting provider or vendor without a rewrite.
Evidence before adoption. New technology is proven on internal work first. It reaches client systems once we understand its failure modes.
Working with an established stack?
Most of our engagements extend existing systems. Tell us what you run today and where it is under strain.