UX and UI design
Screens and flows designed before any code is written. Redrawing a screen is far cheaper than rebuilding a finished app.
TECHNOLOGY
What we work with
We do not start with "what shall we write it in" but with "what does it need to do". The technology follows from the answer — and it is always something we can maintain for years and hand over to another team.
Frontend — what the user sees
- TypeScript
- Vue
- Nuxt
- React
- Next.js
- Angular
TypeScript
JavaScript with type checking. It catches a whole class of bugs while you write, not at the customer.
Vue and Nuxt
Our default for sites and apps where load speed matters and search engines need to find the page.
React and Next.js
When you already have a React team or codebase, we build on it rather than rewriting from scratch.
Angular
For large enterprise applications with a long life, where strict structure pays off.
AI and language models
- Claude (Anthropic)
- ChatGPT (OpenAI)
Claude (Anthropic)
Our primary model for development and AI features — generating code and copy with a focus on quality and reliability.
ChatGPT (OpenAI)
GPT models for content generation, assistant features and integrations where they fit best.
RAG and vector search
AI answers based on your own data and documents, not on guesswork.
AI integrations and automation
Putting AI into real workflows — from chatbots to document and data processing.
Backend — logic and data
- PHP
- Symfony
- Laravel
- Node.js
- Python
- Django
- PostgreSQL
- Elasticsearch
- Redis
PHP — Symfony, Laravel, Nette
A proven base for e-commerce and business systems. We reach for Symfony when the solution has to grow for years.
Node.js
One language across frontend and backend, so the team stops context-switching and code gets shared.
Python — Django
Data processing, integrations and anything AI-adjacent, where Python has the strongest tooling.
PostgreSQL
Where your data lives. We choose it for reliability and for how well it grows with you.
Elasticsearch
Search across a large catalogue that returns in milliseconds and copes with typos.
Redis
A cache layer. Repeated queries are not recomputed, so the site stays fast under a traffic spike.
Mobile applications
- Swift
- Kotlin
- React
- Firebase
iOS — Swift
A native app for iPhone and iPad when you need full performance and access to device features.
Android — Kotlin
The same for Android. We go native where performance or device hardware is the point.
React Native
One app for both platforms. It saves time and budget when you do not need strictly native behaviour.
Firebase
Sign-in, notifications and data sync without standing up your own server.
Infrastructure and operations
- Docker
- Kubernetes
- Terraform
- Ansible
- Prometheus
- Grafana
Docker
The app runs the same on a developer laptop and on the server.
Kubernetes
Cluster design and management. Load arrives and the app scales itself up; load drops and it scales back down so you stop paying for idle.
Amazon Web Services
Cloud design, management and advice — including how not to pay for services you do not use.
Terraform and Ansible
Infrastructure described in code. A server can be rebuilt identically at any time, not by hand and from memory.
CI/CD pipelines
Automated tests and deployment. A change reaches production only once it has passed the checks.
Prometheus and Grafana
Monitoring. We hear about an outage from an alert, not from a call by your customer.
Not sure what you need?
Tell us what the solution has to do. We will propose the technology — and explain why that one.