Dedicated workstation on site
We install the on-premise AI workstation and configure the local open-weight LLMs on your LAN. Procurement runs through MePA/Consip.
IntelliPA
From the workstation on site to signature in 5 steps. IntelliPA generates and verifies administrative acts locally; the signature and the responsibility stay with the RUP.
The 5 steps01 02 03 04 05
RAG over your documents
We index the organisation’s regulations, resolutions, tender specifications and forms. The knowledge base stays on site and is refreshed on an agreed schedule.
We configure the offices
We set up templates and knowledge bases per office (Tenders, Records, Civil Registry, Tax, Urban Planning) and open access via Open WebUI, for 3 to 8 operators.
They generate, query, verify
Operators generate drafts of determinations, resolutions and specifications and query the rules. The reviewing agent flags inconsistencies and outdated references before signature.
The RUP verifies and signs
The final decision stays human: IntelliPA prepares and verifies the draft, the RUP (the responsible officer) or the manager validates, signs and takes responsibility for the act.
Under the hood
Local AI grounded in the organisation’s documents
A pipeline of AI components, each with a job: index the organisation’s documents, retrieve the relevant passages (RAG), generate the draft on your template, pass it to the reviewing agent and cite the sources. Answers come from the organisation’s regulations, resolutions and rules, not from the memory of a model trained elsewhere. Everything runs on the workstation on site: documents and requests never leave the organisation, and no text is sent to the cloud.
How we keep control
Guarantees built into the product
Answers with the source
Every answer and every draft cites the organisation’s documents it comes from (a regulation, a resolution, an article of the code): at review you verify, you do not rebuild.
Data stays inside the organisation
Models, knowledge base and processing run on the workstation on site, on your LAN. No document and no request leave for the cloud.
The signature stays with the RUP
IntelliPA generates and verifies, but signs nothing. The decision, the signature and the responsibility for the act stay with the RUP or the manager.
No training on your data
The LLMs are open weight and run locally. The organisation’s documents serve to answer your requests, not to train shared models.
Where IntelliPA sits
Compared to other approaches
| Approach | What it does | Where the data sits | Limit |
|---|---|---|---|
| Public cloud AI chatbot | Generates generic text | In the vendor’s cloud | Documents leave the organisation, no knowledge base on the acts, no consistency check |
| Copy and adapt a previous act | Manual reuse of a previous act as a template | Inside the organisation (manual) | Slow, propagates errors and outdated references, depends on the officer’s experience |
| External consultant or firm | Specialist drafting per project | With the external consultant | Costly, not always available, does not scale to daily volume |
| Cloud SaaS AI for the public sector | Vertical AI assistant | In the vendor’s cloud | Data leaves the organisation, depends on connectivity, recurring per-user fee |
| IntelliPA | Generative AI on site: generates, verifies, cites sources | On-premise, inside the organisation | Requires the dedicated workstation on site |