AI explained

RAG explained, part 4: tools, costs and a realistic way in

The software for RAG usually costs nothing. What you pay for is computing power and, above all, working time – this part sorts out what is available and which items really do add up.

  • 7 min read
Read article →
AI explained

RAG explained, part 3: when does the effort pay off?

The question is often asked with reference to page counts or document counts – and that is what makes it misleading. What matters is less the volume than how well ordered your documents are.

  • 6 min read
Read article →
AI explained

RAG explained, part 2: not every RAG needs a vector database

Read up on the subject and you meet embeddings and vector databases everywhere – so often that the two look like the same thing. They are not: looking things up works in three ways, and not every one of them needs a vector database.

  • 5 min read
Read article →
AI explained

RAG explained, part 1: when AI needs to know your own documents

An AI assistant knows a great deal about the world and nothing about your business. RAG connects the two – the AI looks things up in your documents before it answers.

  • 6 min read
Read article →
Practical tip

From enquiry to finished offer – how a team of AI agents works at ValBit

At ValBit, a team of AI agents drafts our offers – anonymised, reviewed and approved by a human. A look behind the scenes of our own working day.

  • 5 min read
Read article →
About us

Hello! We are ValBit – and this is what we do

To kick off our blog, we introduce ourselves: who is behind ValBit, what we do for companies, public administrations and private individuals – and what we will write about here.

  • 4 min read
Read article →

Notes on our blog articles

A topic you would like to read about?

Tell us what we should write about – we are glad to pick up questions from practice.

Suggest a topic