RUNNING IN PRODUCTION SINCE 2024 · 1.3M CHUNKS

Your documents,
answering back.

A retrieval assistant over everything your organisation has ever written — multilingual, self-hosted, and it cites its sources on every answer. This is a real system I run in production; this page is the Aurora look from my Showroom.

Build one for my team See the production system →

Features

Answers you can check, not just believe.

[01]

Every answer, cited

Responses come with pointers into your own documents — the receipt is part of the reply. Wrong answers become findable; right ones become trustworthy.

[02]

Reads what your team reads

Multilingual embeddings mean the archive answers in the language you ask — the reference build indexes sources from 50+ outlets across the world.

[03]

Yours, on your machines

Self-hosted retrieval — your documents never leave your infrastructure. Pair with any frontier model, or none at all.

How it works

Three moves, then it's just there.

Ingest

Your archive — PDFs, wikis, mail, databases — is cleaned, chunked, and embedded by pipelines built for your formats.

Index

Vectors go into FAISS on your hardware. The reference system holds 1.3 million chunks and answers in milliseconds.

Ask

A chat surface your team actually adopts — every claim linked back to the paragraph it came from.

0document chunks in the reference index
0articles curated into the corpus
0global outlets, many languages
0trackers — here and in what I build for you

Engagements

Scoped like software, quoted like a craftsman.

Pilot

Prove it on your data

Fixed written quote within 48 hours of a free scoping call

  • One document collection, ingested end-to-end
  • Working assistant your team can try in weeks
  • Evaluation report — where it's strong, where it isn't
Start a pilot
Production

The full system

Fixed written quote within 48 hours of a free scoping call

  • Full-archive pipelines, scheduled and monitored
  • Citations, permissions, and your model of choice
  • Documented handover — you own every line
Commission it

FAQ

Asked, answered.

Is this a product I subscribe to?

No — it's a system I build for you and hand over. The reference build has run in production since 2024; yours would be shaped to your archive, your privacy rules, and your stack.

Our documents are sensitive. Where do they go?

Nowhere. Ingestion, the index, and retrieval all run on infrastructure you control. If a frontier model is used for drafting answers, you choose exactly what context it may see.

How big does our archive need to be?

The architecture scales down gracefully — a few thousand documents is a small, affordable build, and the same design grows past a million chunks without being rebuilt.

What does it cost?

By written estimate after a free scoping call — a fixed quote within 48 hours, no meter running. Full terms live on the rate card.

Why does this page look nothing like the rest of your site?

Because it's a Showroom look — a demonstration that the house style is a choice, not a limitation. Same hand-written code, same zero-tracker guarantee, different costume.

Ask your archive anything.

Book the free scoping call