Internal tools
Admin panels, ticketing systems, calculators, reports. Applications that automate your team’s daily work.
03 / Software
When there is no ready-made tool on the market, or the existing one does not fit your processes, we build it from scratch: from internal tools to complex systems and APIs. We work in a modern way, with AI in the development process, and treat security as part of the project from the first line of code.
Admin panels, ticketing systems, calculators, reports. Applications that automate your team’s daily work.
Connecting systems, exposing secure interfaces, synchronizing data between applications.
Customer portals, platforms with login and roles, applications with their own business logic.
Semantic search, document classification, assistants based on your data: where AI genuinely helps.
Taking over existing code, audit, refactoring, adding tests and automated deployment.
A quick, working seed of a product to test an idea with real users.
We gather requirements, design the architecture and split the project into stages, each delivering a working result.
Business logic, databases, interfaces for your own and external systems.
Clear web applications tailored to the work of the people who will use them.
Authentication and permissions, data protection, secrets management, review for common vulnerabilities.
Unit and integration tests that protect against regressions with every change.
Every change goes through review and automated checks, and deployment is repeatable and reversible.
Launch on your infrastructure, monitoring, backups, further development.
A description of how it works, deployment instructions and short training for your team.
What it looks like from the technical side: the key elements we take into account in custom projects.
REST and OpenAPI (Swagger)
A described, versioned API with documentation that can be generated and tested.
Webhooks and integrations
Notifications between systems, retries, request signing.
Queues and background jobs
Asynchronous processing of longer operations (e.g. RabbitMQ, Redis).
GraphQL (when it makes sense)
Flexible queries for complex interfaces.
OAuth2 / OIDC, JWT
Login, tokens, session refresh, integration with Keycloak or company SSO.
RBAC
Roles and permissions matched to your company’s structure.
Secrets management
Passwords and keys kept out of the code, rotation.
PostgreSQL: schema and migrations
Data modelling, zero-downtime migrations, indexes and transactions.
Cache (Redis)
Faster reads and lower database load.
Data import and export
Moving data from Excel, CSV and other systems.
Unit, integration and e2e tests
Tests that protect against regressions, run on every change.
Code review
Every change is reviewed by a second person.
Dependency and code scanning
Detecting vulnerabilities in libraries and code (e.g. SonarQube, Trivy).
OWASP Top 10
Designing with common vulnerabilities in mind: injection, XSS, authorization flaws.
Docker and CI/CD
Repeatable image builds and deployment, the ability to quickly roll back changes.
Logs, metrics and health checks
Application observability (Prometheus, Grafana, Sentry).
12-factor
Configuration through the environment, stateless processes, easy scaling.
If you are trying to set this up yourself and got stuck on one of the points above, tell us where you are. We will point you in the right direction, or do it for you.
We talk about the process and the business goal before we start talking about technology.
We split the work into stages with a clear scope. You receive an estimate and a schedule.
We show the first working versions early, so corrections happen along the way, not at the end.
We test with you, deploy and hand over the documentation.
Ongoing care and development as needed.
AI tools speed up developers’ work: code drafts, tests, documentation. A human is still responsible for quality and security: every change goes through code review and tests.
We agree on the rights to the code and the terms of its use in the contract, before work begins.
Yes. We start with a review of the code and deployment, then propose an order of priorities.
After the analysis we split the project into stages and price each one. That way you know what you are paying for and can stop the project after any stage.
Yes, that is often the best approach: a small MVP or a single module that lets you test the collaboration and the idea.