Multi-tenant SaaS platform
One product for many customers with separate data, roles and per-customer billing, designed to grow without rewriting.
Custom software and cloud platforms, one sheet at a time.
Native iOS and Android mobile apps, multi-tenant SaaS platforms, Computer Vision and AI systems, Data Governance, GIS, industrial management systems. Custom development and architecture, without compromise.
Every platform is a stack of thin sheets.
Separating them helps to build them, manage them and grow them one at a time, with continuous delivery and pinned versions.
Web interfaces and apps built on the services, with in-house libraries for data access and declarative interfaces.
Service marketplace, access control, request rate limiting and load balancing.
Independent microservices, one per function, that talk to each other through APIs.
PostgreSQL with a dedicated schema for each client and replication to the standby node.
Containers, reverse proxy and private network on public cloud and local nodes, described as code.
Why it matters.
A platform grows for years: the decisions taken at the start weigh more than the code written afterwards. That is why we start from architecture: domain boundaries, data, integrations and the security model.
We build to be replaceable: open stacks, documented interfaces, infrastructure described as code. The customer can evolve the system alone, or with us.
How it works, layer by layer.
Five thin sheets, one on top of the other.
- Applications
- Web interfaces and native apps, in SwiftUI and Kotlin, built on the services with our own libraries for declarative interfaces.
- Platform
- Service marketplace, access control, request rate limiting and load balancing.
- Services
- Independent FastAPI microservices, one per function, communicating through APIs.
- Data
- PostgreSQL with one schema per customer, replication to a standby node and partitions for historical data.
- Infrastructure
- Containers with a reverse proxy and private network, described as code and deployed with release pipelines on pinned versions.
What we build. From strategy to production.
Four families of work, one method: architecture first, then code.
- Native mobile applications
- iOS and Android apps designed for real users, with high-performance native interfaces integrated with company systems. From prototyping to store publication.
- Cloud and SaaS platforms
- Scalable multi-tenant architectures, designed to grow without rewriting. Robust APIs, structured data, monitored container deployment. Nebula, our cloud platform, is built this way.
- Applied AI and Computer Vision
- Computer vision for quality control, language models for document automation, real-time data pipelines, Edge AI.
- Industrial systems, IoT and GIS
- From sensor integration to territory management: industrial protocols, ERP and SCADA, operational dashboards.
How we work.
- 01
Architecture before code
Every project starts with an in-depth technical analysis, not with a quote. We design the whole system, with data flows, domain boundaries and integrations, before writing a single line of code.
- 02
Reasoned technology
No arbitrary choices. Every component has a documented reason: open, replaceable stacks with no proprietary lock-in. The technology choice serves the problem, not the other way round.
- 03
Iterative construction
Short cycles, continuous feedback. Every iteration produces working, tested and documented code. The client sees real progress, not progress slides.
- 04
Guaranteed reversibility
We deliver systems that clients can evolve on their own. Complete documentation, clean code, zero opaque dependencies. The system truly belongs to you.
- 05
Structured continuity
Post-delivery support is planned and documented from the start, not added as an afterthought. We don't abandon systems after the first release.
Use cases.
Native mobile app
An iOS and Android app with a polished interface, integrated with company systems and published on the stores.
Document automation with language models
A service that reads documents, extracts the data and records it in the management system, with human review where needed.
Data, integrations and standards.
We develop with open-source languages and frameworks and documented APIs. Code, documentation and release procedures pass to the customer, with no opaque dependencies.
Privacy, security and compliance.
- Customer isolation
- One database schema for each customer and access controls on every service.
- Traceable releases
- Every version in production is identified and reproducible, with automated checks before release.
- GDPR and AI Act
- Functions that process personal data or use artificial intelligence models are designed to comply with the GDPR and the AI Act.
Words worth knowing.
- Multi-tenant
- A single platform that serves several customers, with data and settings kept separate for each.
- Microservice
- A small, independent service, responsible for a single function and reachable through APIs.
- Infrastructure as code
- Servers, networks and containers described in versioned files, so the environment can be recreated identically.
- Lock-in
- Dependence on a supplier or a proprietary technology that makes switching costly.
Frequently asked questions.
What technologies do you use?
We use open stacks: Python and FastAPI, Java and Spring Boot, PostgreSQL, Kotlin and SwiftUI for apps. We choose according to the problem, not the other way round.
Does the code stay with us?
Yes. We deliver code, documentation and release procedures: the system can evolve independently, with no proprietary dependencies.
Do you also develop mobile apps?
Yes: native iOS and Android apps, from prototyping to publication, with the same architecture as the services.
The other areas.
People
Museums, kiosks and proximity marketing: tools that speak to people in the right place.
Open the pagePlaces
Indoor and outdoor localization, augmented reality search and contextual digital signage.
Open the pageTerritory
Multi-tenant GIS for public bodies: maps, open data and aggregated flows of people.
Open the pageProcesses
Predictive maintenance and statistical process control on machines and lines.
Open the pageSensors
IoT and Edge AI: models on the device and nodes that coordinate.
Open the pageDo you have a platform to build or to grow?
Tell us about the project: within 48 hours we tell you whether we can help and with which stack.
Let's talk