Skip to main content
Home Services

Obsigen AI

European AI designed with data privacy in mind — built by Visera.

Dedicated infrastructure · Processing in the EEA · Zero Data Retention

interface simulation · sample data

One instruction, finished files.

Obsigen reads, computes, writes and runs code — right in the browser. For every task, Obsigen GRID selects the right models, run by providers in the EEA. Your data stays on dedicated infrastructure; the model receives only what the task requires.

01
Finished results

Documents, spreadsheets, presentations and graphics — not just chat answers.

02
Dedicated tenant

Your own database and your own storage — on infrastructure assigned exclusively to you.

03
Verified results

Tests decide, not model agreement — in Titan mode a task’s result is verified before it reaches the user.

Obsigen GRID: several models, one verified result.

In the default Titan mode, a task goes through the full GRID: model selection, collaboration of several models, evidence-based verification and the output gate. The result reaches the user only after verification. Simple questions such as “hello” are answered straight away by a single model.

REQUEST MODEL SELECTION COLLABORATION n = 4 inference points VERIFICATION evidence-based GATE verified only DONE ONE TURN · ORCHESTRATION FIELD GRID KEEPS NO TASK CONTENT
Normal

Questions, short tasks and work on files.

Genius

Harder reasoning and long documents, with basic GRID orchestration.

Titan · default

Reports, analyses and code with tests — full GRID orchestration and a plan preview.

One system. Three interfaces.

The same Obsigen, where your team already works.

Obsigen in the browser: a task with steps for reading a CSV file and running the analysis in a sandbox, a checklist and the finished DOCX report; Workspace panel with files on the right.
product screenshot
Obsigen in the browser

A web application for working with company documents and data. Nothing to install. Reads files, calculates in spreadsheets, writes documents and decks; finished files land in the Workspace panel.

interface simulation · sample data
Obsigen Forge for VS Code

A coding agent in VS Code — from spec to tested code. Syntax checked after every save; commands are classified before they run: reads without asking, writes and deletions with consent, the rest blocked.

interface simulation · sample data
Obsigen Runtime

An agent on the workstation, no IDE. It works on local files and saves results to the folder you choose; consent for commands comes in a native system dialog. macOS and Debian; on Windows in preparation.

What we're building next.

The Obsigen team works on the product every day. The next release of GRID is already in the works.

coming soon

Model selection and verification are the first layer of GRID. We are developing the next one - the Artifact GRID release - as a shared, verifiable state of knowledge about the task: models add hypotheses, artifacts and test results to it through references, without passing content to one another. Every conclusion has a provenance, and verification decides what is true. Details will be published on launch day.

ARTIFACT GRID · STATE OF KNOWLEDGE IN YOUR TENANT MODEL α MODEL β MODEL γ TEST · RUN art·0912 specification.pdf#p12 reference art·0417 orders.ts#L184 supported run·0044 tests: passed verified wn·07 counterexample falsified wn·08 · wn·09 depend on wn·07 retracted ref art·0912 → node (1,2,2)run·0044 ✓ → wn·03 verifiedwn·07 ✕ → retraction: wn·08, wn·09

Visualization · artifact codes are illustrative

Deployment models

Data always on dedicated infrastructure. You can change the model later.

Managed Cloud

We deploy and maintain Obsigen on dedicated OVHcloud infrastructure in the EEA. One contract: infrastructure, deployment and maintenance.

Best for the fastest start, with no IT effort.

Self-hosted

We install and maintain Obsigen in your environment. Hosting, access and security policies stay under your control.

Best for organizations with their own IT and hosting.

On-premise · on request

Data, GRID and models in your own data center, with no connections to external model providers.

Best for regulated sectors and highly sensitive data.

How Visera deploys Obsigen

Obsigen is our product. We deploy it, connect it to your documents and systems, and keep it running.

1 · Discovery

Goals, data, systems and constraints — and the deployment model that fits.

2 · Deployment

Your dedicated tenant, access for the team and integrations agreed before we start.

3 · Team start

Working sessions on your own documents and processes, not on generic examples.

4 · Maintenance

Updates and support from the team that builds the product.

See Obsigen on your own example

A 45-minute online demo on a document, spreadsheet or process you bring.