AI training
Building RAG and Knowledge Assistants
Build a question-answering assistant over your own documents, with citations, access control and quality testing.
Enquire about this course- Duration
- 2 days
- Level
- Intermediate
- Delivery
- Live online or on-site, for individuals or private teams
- Prerequisites
- Comfortable programming in at least one language.
- Certificate
- Not included
- Dates and price
- Contact us for current dates and pricing
What you will build
A knowledge assistant that answers from a document set, cites its sources and respects who may see what.
What is included
- Hands-on labs with worked solutions
- Course materials and sample code
- Follow-up questions answered after the course
Course outline
What you will cover
Day 1: Getting the content right
- How retrieval-augmented generation works
- Document preparation and chunking
- Embeddings and vector search
- Hybrid retrieval
- Lab: index a document set and search it
Day 2: Answers you can trust
- Prompting with retrieved context
- Citations and the option to say "I do not know"
- Permissions and guardrails
- Evaluation sets and monitoring
- Lab: complete the assistant and test its quality
Who it is for
- Developers and data engineers
- Teams building an internal knowledge assistant
What you will be able to do
- Ingest, chunk and index documents for search
- Combine keyword and meaning-based retrieval
- Return answers with citations and respect permissions
- Measure and improve answer quality
Tools
Related courses
Generative AI for Developers
Add language-model features to real applications: prompts, structured output, retrieval, evaluation and cost control.
Course details →Azure AI Engineering
Build AI solutions on Azure: language models, search, document intelligence and secure deployment.
Course details →Agentic AI Bootcamp: Build Working AI Agents
A six-week, hands-on programme in which you build and deploy seven working AI agents, from research assistants and support bots to data pipelines.
Course details →