Live answer, requirement 1

Retrieval-augmented generation

Grounded, cited answers over the customer's own documents, with access rules applied before retrieval.

The requirementBuild retrieval-augmented generation over a customer's own documents, so answers are grounded in their sources and cite them.

Try it

Ask a question about the engineering work Gus IT has published. The demo retrieves from the 14 use cases on gusit.de with the hybrid method above and shows every score it used.

How it works

Scope the sourcesSit with the data owner, list the repositories, and agree what may be indexed and who may see it. Access rules are decided before any document is embedded.
Chunk and embed inside the boundaryDocuments are split into passages and embedded. Luca runs embeddings on a local GPU by default, so customer text does not leave the environment; a hosted embedding model is the fallback.
Hybrid retrievalEach passage is scored by semantic similarity blended with keyword overlap, then Maximal Marginal Relevance picks a diverse set, at most two passages per document, so one repeated paragraph cannot crowd out the rest.
Grounded generationThe model answers only from the retrieved passages, cites them by number, and says so when nothing clears the relevance threshold instead of guessing.

Tools

Tool	What it does here	Status
Luca Express RAG engine	Document store, embedding and query API inside the Luca workspace.	Available
Luca On	Embeddable website chat answering from a per-customer knowledge base.	Available
Claude with MCP	Connects the assistant to customer systems through open Model Context Protocol connectors.	Available
Cloudflare Workers AI	Runs this public demo at the edge: bge-base embeddings and a Llama instruct model.	Demo only

The full write-up covers access control, evaluation and how this runs inside a customer environment.

Read it on gusit.de

Retrieval-augmented generation

Try it

How it works

Tools

Other answers