# The Technology We Build With, and Why

We run deep on .NET and React and pick the right tool for each job rather than forcing one stack onto every project. Here is what we use at each layer and where it fits.

ON THIS PAGE

- Stack at a glance

- Explore by layer

- Front end: React, Next.js and TypeScript

- Back end: ASP.NET Core and Node.js

- Data: PostgreSQL and pgvector

- Cloud and delivery

- Mobile: cross-platform or native

- AI: models behind a stable interface

- How we choose

## Stack at a glance

## Explore by layer

- REACT, NEXT.JS, TYPESCRIPT Front end Web applications, dashboards, marketing sites. Web development

- ASP.NET CORE, NODE.JS, REST APIS Back end Business logic, integrations, versioned APIs. SaaS development

- POSTGRESQL, PGVECTOR Data Relational data, search, embeddings. AI agents and RAG

- AWS, AZURE, GCP Cloud Hosting, storage, queues, CI/CD. Cloud and DevOps

- REACT NATIVE, .NET MAUI, NATIVE Mobile iOS, Android, Windows and Mac apps. Mobile apps

- LLMS SUCH AS GEMINI, RETRIEVAL, AGENTS AI Assistants, document analysis, automation. AI services

## Front end: React, Next.js and TypeScript

TypeScript gives the front end the same safety net as the back end, and shared types mean an API change shows up as a compile error instead of a production bug. We use React for application interfaces and Next.js where server rendering or static generation helps performance and search visibility. Details on the web development page.

## Back end: ASP.NET Core and Node.js

ASP.NET Core is our default for systems with rich business rules, long lifetimes and strict performance needs. Node.js suits lightweight services, background workers and teams already invested in JavaScript. Both expose versioned REST APIs with typed contracts, authentication and role-based authorisation.

## Data: PostgreSQL and pgvector

PostgreSQL covers most needs with one dependable engine. With the pgvector extension it also stores embeddings, so retrieval for AI features can sit next to your transactional data under the same access rules instead of in a separate system. See AI agents and RAG systems.

## Cloud and delivery

We deploy to AWS, Azure or GCP depending on where you already are and what your compliance needs say. Infrastructure is defined as code, builds and tests run in CI, and releases are scripted so that deploying is boring. More on cloud and DevOps.

## Mobile: cross-platform or native

React Native lets one team ship to iOS and Android from a shared codebase. .NET MAUI suits teams that want to share code with a .NET back end and target Windows and Mac too. Where performance or platform features demand it, we go native. See mobile app development.

## AI: models behind a stable interface

Model calls run behind typed back-end services, outputs are validated, and failure modes such as timeouts and refusals are handled explicitly. We choose models per use case, including Gemini, behind an internal interface so the model can change without rewriting the product.

## How we choose

- Fit with the problem and your team, before preference.

- Maturity, community and long-term support.

- Ability to hire for or hand over to the people who will maintain it.

- Operating cost, including hosting, licences and on-call effort.

A scoping sprint includes an architecture and technology recommendation you can use with any team.

## Have a project in mind?

Tell us what you are building. A senior engineer replies within one business day.
