# RAG development and custom AI chatbots

> RAG development and custom AI chatbots that answer from your company documents, with sources. For internal knowledge search, support and your website.

- Provider: Prayag Dalal, ShipFast (https://shipfast.online)
- Pricing: Fixed quote after a 30-minute call, or $30/hour
- Timeline: Set in your written scope
- Canonical: https://shipfast.online/services/rag-development/

**Outcome:** An AI chatbot that answers from your own documents, with sources.

## Who it is for

- Teams whose knowledge is scattered across docs, PDFs and wikis
- Support teams answering the same questions every day
- Companies that want an AI assistant without sending everything to a black box

## What you get

- Ingestion of your documents with a plan for keeping them up to date
- Retrieval tuned on your real questions, with cited sources in every answer
- A chat interface for your team or your website
- An evaluation set so answer quality can be measured, not guessed
- Clear notes on costs, data handling and limits

## How a RAG chatbot works

Your documents are split into passages, indexed for semantic search, and searched for each question. The LLM answers only from what it finds and links back to the source, so people can check it. See the full pipeline in the diagram below.

## RAG or fine-tuning?

For most company chatbots, RAG is the right choice: your documents change, people need sources, and nothing has to be retrained when you add a file. Fine-tuning is for teaching a model a format or tone, not for storing facts. More in [RAG or fine-tuning?](https://shipfast.online/blog/rag-vs-fine-tuning/)

## Built to be measured

Before launch we collect real questions from your team or customers and use them as a test set. Changes to prompts, models or documents are checked against it, so quality goes up rather than drifting.

## FAQ

### Will it make things up?

Every answer is grounded in retrieved passages and shows its sources, and we test against a set of your real questions. When the documents don't contain the answer, it says so.

### Do you need to train a model on our data?

Usually not. A RAG chatbot reads your documents at question time, so there is no model training and documents can be added or removed at any time.

### Can the chatbot go on our website?

Yes. The same assistant can serve your team internally or your customers on your website, with different document sets for each.

### Which LLM do you use?

It's chosen per project based on answer quality, cost and your data rules, typically Claude (Anthropic) or GPT (OpenAI).

---

Show me the documents your team keeps searching: https://shipfast.online/contact/?service=rag-development or book a call at https://cal.com/prayagdalal/30min