Highlights
- Explain RAG versus fine-tuning tradeoffs
- Chunk and prepare documents for embedding
- Generate and store vector embeddings
- Set up and query a vector database
- Build a retrieval pipeline
- Construct prompts with injected context
- Measure retrieval quality
- Handle conflicting or outdated documents
- Connect to a real document store
- Deploy the pipeline as an API
Course Details
RAG versus fine-tuning:
cost, latency, accuracy, and maintenance tradeoffs shown with real benchmark numbers
Document preparation lab:
splitting strategies, cleaning noise, and handling PDFs, Word docs, and HTML in one pipeline
Embedding models compared:
OpenAI ada, open-source alternatives, and a decision framework for choosing based on your data
Vector database setup:
schema design, indexing choices, and running your first similarity search query against real documents
Evaluation lab:
building a test set and scoring retrieval precision, recall, and final answer quality with a repeatable method
Retrieval pipeline build:
query embedding, nearest-neighbour search, and result ranking working together end to end
Prompt engineering for RAG:
context injection patterns that demonstrably reduce hallucination in tested examples
Conflict resolution:
practical strategies for versioned documents, contradictory sources, and stale content in the knowledge base
Packaging and deployment:
wrapping the completed pipeline as a callable API and running it in a cloud environment
Enterprise connector lab:
pulling live documents from SharePoint, Confluence, or an internal file store into the pipeline
Who should attend
|
Developers and Engineers |
Feedback
4.8 out of 5 average
"Our tailored course provided a well rounded introduction and also covered some intermediate level topics that we needed to know. Clive gave us some best practice ideas and tips to take away. Fast paced but the instructor never lost any of the delegates"
Brian Leek, Data Analyst, May 2022
