Swiss AI Platform
The overview: how control, execution, models, compute nodes and the shared data core work together.
View details
The technical platform behind our AI solutions.
Our solutions run on our own platform. It manages access, tasks, AI models and operating locations. It has three parts: Swiss AI Studio (control), Swiss AI Agents (execution) and Swiss AI Factory (models). Computing power comes from compute nodes that can be expanded.
In brief: Swiss AI Labs is the technical arm of Swiss AI Agency & Labs. We develop and run the platform on which our products and client solutions run. AI models, compute nodes and operating location (your own premises, a Swiss cloud, a combination) are designed to be interchangeable.
Platform building blocks
Three layers, clear roles.
Swiss AI Studio
Control: settings, models, agents, knowledge, access, status, monitoring and approvals in one place.
View detailsSwiss AI Agents
Execution: AI agents take on tasks, use approved tools, start workflows and request approvals.
View detailsSwiss AI Factory
Models: every task goes to the right AI model, with load balancing, a fallback model if one fails and quality checks – local or external.
View details
Architecture
Eight layers, shown publicly.
Principle, layers, modularity and data flow at a high level – without trade secrets. Applications at the top, operating models at the bottom. In between: the shared data core, control, agents, model access and compute nodes.
Layer 1Applications
MABEX, Call Agent, Easy Factura, Meeting Intelligence, Offertwerk and custom client solutions
Layer 2Industry modules
Real estate, trades & building services, accounting & fiduciary, healthcare, manufacturing and finance
Layer 3Shared data core (Shared Business Core)
Customers, companies, people, projects, documents, users, roles, permissions and master data
Layer 4Swiss AI Studio
Control, configuration, oversight and monitoring
Layer 5Swiss AI Agents
Agents, tasks, tools, triggers, orchestration and approvals
Layer 6Swiss AI Factory
Access to all models, model selection per task, capability testing, load balancing, failover and quality checks
Layer 7AI nodes & models
Node 01, Node 02, Node N, local models, specialised models, external models and RAG / knowledge systems
Layer 8Operating models
On-premise, Swiss private cloud, controlled cloud and hybrid
Key features
What defines the platform.
Scalable via compute nodes
You can expand computing power by adding compute nodes, with load balancing and local specialisation. We plan and test every expansion.
View detailsInterchangeable models
Application and AI model are separate. You can replace one model with another or add a local model alongside a cloud model. Testing is part of it.
View detailsThe right model for each task
Not every model can do everything. Swiss AI Factory assigns tasks by capability, data protection, speed, cost and availability.
View detailsOn premises, in the cloud or combined
On your own servers, in a Swiss private cloud, with controlled external models or a mix: the operating location follows the nature of the data.
View details
Architecture & operations
Six topics, six detail pages.
Private AI
AI systems under your own control: data, models and operation where you decide.
View detailsOn-premise & hybrid
Running on your own servers, in the cloud or combined: comparison and prerequisites.
View detailsModel Routing
The right AI model for each task instead of one model for everything.
View detailsRAG Engineering
Answers from your own documents, with sources and permissions.
View detailsAI infrastructure
Compute nodes, GPUs, network and storage for local, private and hybrid systems.
View detailsAI Operations
Operation: measure, monitor, approve, control costs.
View details
Demo vs. operation
What a solution in production also needs.
A demo needs a model and an input. A solution in production needs data integration, permissions, approvals, monitoring, fallback rules, cost control and support.
That is exactly the part the Labs take care of – for our products and for custom client solutions.
Case studies
Technical case studies.
We do not present architecture examples without a client as case studies.
Frequently asked questions
Frequently asked questions about the Labs
Is the Swiss AI Platform an operating system?
We call it a platform, control layer or orchestration layer: a modular control and operations layer for business AI.
Which models do you use?
Local, specialised and external models – depending on the task, data protection, speed, cost and availability. Swiss AI Factory handles the assignment.
What happens if a model is no longer available?
The architecture is designed to largely decouple models and infrastructure from the application layer. Switching models should not mean reinventing the business application.
Do you show the internal architecture?
Publicly, we show the principle, layers and data flow at a high level. Internal topology, routing rules and security details stay internal. Even in client meetings, we only discuss them under confidentiality.
Next step
For AI solutions where data, infrastructure and control are critical.
No-obligation introductory call about architecture, data, models and operation.