15-day free trialStart XGEN right in your browser — nothing to installStart free trial

Solutions · Customer cases

Public Sector customer cases

Customer cases applying AI to public-sector work — procurement, search, and citizen services

01Overview

AI for public sector work

Public-sector work deals with large volumes of regulations, notices and civil-service documents, and the grounds behind every answer matter. Network separation and security certification requirements apply too. XGEN is applied by searching and answering from documents inside the internal network, and holds GS certification grade 1.

02Challenges

What public sector teams run into

  1. 01

    You must know the exact term to find it

    Procurement and notice search only works if you know the exact terms and categories, so results vary by person.

  2. 02

    Answers need grounds

    Guidance and decisions in public work must show which regulation or document they rest on.

  3. 03

    Security requirements come first

    Without internal-network operation, access control and privacy protection, adoption rarely gets past review.

03Approach

How it is applied, step by step

  1. 1

    Internal-network deployment

    XGEN and the LLM are installed on the institution's servers and configured to run without external connections.

  2. 2

    Meaning-based search

    Query intent is interpreted to find related notices and regulations, presented with source documents.

  3. 3

    Access control and de-identification

    Visible documents are scoped by user permission, and personal data is masked.

  4. 4

    Operating records

    Queries, answers and actions are kept in audit logs for operational review.

04Deployments

What we have delivered

A public mutual-aid association|In operationXGEN

Large-scale procurement search, rebuilt on AI

A public-sector engagement rebuilding a mutual-aid association's procurement search on AI, then stabilizing it through on-site operations

Challenge

Search had to be modernized so that accurate results came back from a very large body of procurement data.

What we built

The procurement search layer was rebuilt on AI search technology, and the service was stabilized through an on-site operating team.

Scope delivered

  • Procurement search layer rebuilt
  • Handles large-scale data
  • On-site operations
Read the full case

05Typical work

  • 01Meaning-based procurement and notice search
  • 02Q&A over rules and guidelines with cited grounds
  • 03Assisting with drafts for civil-service responses

06Before you adopt

  • Installation and external-integration scope vary with each institution's security guidelines.
  • GS certification applies to the product version and test scope at the time of certification.
  • Search quality is affected by how current and well-organized the documents are.
Public Sector customer cases · Plateer AI Labs