From the NVIDIA edition of October 27, 2026
NVIDIA's French public-service assistant code ships six sourced procedures and ten test prompts, and states tuning does not add facts
NVIDIA (NVIDIA-NeMo/Nemotron repository) · Government citizen services (open-source reference code) · France; Global
- Publisher
- GitHub: NVIDIA-NeMo/Nemotron
- Original publication
- Undated repository README (retrieved 2026-10-27)
- Source retrieved
- 2026-10-28
What happened
The reference implementation contains six procedures whose facts come from published pages, each with source URL and verification date, plus ten test prompts with expected behavior. The README says fine-tuning teaches behavior such as staying inside supplied facts but does not add facts the model lacks, and that production assistants should read fees, documents and deadlines from the agency's own data or retrieval at answer time. It says the workflow has been run end to end on a DGX Spark.
Why it matters
Gives agencies a concrete template for sourcing and dating facts and an explicit vendor statement that fine-tuning is not a substitute for authoritative retrieval, which is central to any public-service assistant procurement.
Evidence and measured results
Repository documentation and code structure; ten test prompts is a smoke-test scale, not an evaluation.
Limitations and uncertainty
Six procedures and ten prompts is far smaller than a production service catalog; README is undated and may change.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-10-28; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Agencies expect fine-tuning to teach the model their rules, risking outdated or invented facts.
- Stakeholders
- Program office owner
- Web/content manager
- CIO
- Procurement officer
- Discovery questions
- Where are your authoritative fees and deadlines published?
- Who verifies them and how often?
- Will your RFP require sourced answers?
- Value hypothesis
- A sourced, dated fact store plus behavior tuning yields assistants that can be audited and updated without retraining.
- Potential engagement
- Fact-store and retrieval design engagement for priority services.
- Unsupported claims
- Ten test prompts is not a production evaluation.
- No measured accuracy results in the README.
Pre-sales engineering
Role takeaway
- Fit
- Any citizen-service assistant with volatile facts.
- Architecture and integration
- Tuned model for behavior; retrieval over agency-owned dated sources for facts.
- Prerequisites
- Published procedure pages
- Source URL and verification date per fact
- Retrieval pipeline
- Constraints
- Tuning does not add facts
- Sample scope is six procedures
- Security
- Limit fact store to official content; privacy review for any non-public data.
- Proof of value
- Every answer traceable to a dated source
- Expanded holdout pass rate
Delivery
Role takeaway
- Work
- Build fact store with owners and verification dates; expand procedures in phases.
- Dependencies
- Content owners
- Retrieval infrastructure
- Ownership
- Program office content owners with IT support.
- Skills and adoption
- Train content owners on verification workflow.
- Governance checkpoints
- Fact provenance review
- Re-verification cadence agreed
- Acceptance criteria
- All facts sourced and dated
- Holdout thresholds met
- Risks
- Stale facts
- Under-resourced content maintenance
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Separate behavior tuning from fact delivery: tuned model plus retrieval over agency-owned, dated sources.
Governance
Who approves, reviews and stays accountable for outcomes?
Each fact should carry a source URL and verification date; agencies need a review cadence for re-verifying facts.
Security and privacy
What data, permissions and controls need testing?
Retrieval sources should be official published content; avoid placing resident records in the fact store without privacy review.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Program staff must maintain the fact store; content owners need time allocated for verification.
Procurement
What should contracts, pricing and exit terms secure?
RFPs for citizen-service assistants can require dated, sourced fact stores and retrieval-at-answer-time for volatile facts.
Operating model
Which teams own the service once it runs?
Introduces a content-operations role for fact verification alongside model operations.
Publication history
- 2026-10-27NVIDIA · Issue 5212 resources
Stable resource ID: nvidia-nemo-french-public-service-assistant-repo-scope-20261027