Training: Unleashing search engines potential with RAG
Register
Duration | Price | Location |
---|---|---|
1 day. | 1400 € excl. VAT / person | Station F, 5 Parvis Alan Turing, 75013 Paris or via videoconference |
Objectives
The goal is to enable developers and search engine maintainers to fully power up search engines by implementing a Retrieval Augmented Generation (RAG) solution. RAG combines all the generation and learning capacities of Large Languages Models (LLMs) with the relevance and precision of vector search.
Target
Software developers Search engine maintainers Data engineers An understanding of software development will permit you to get the most of this training course.
All our courses are accessible to people with disabilities.
Curriculum
Introduction to Search engines
- The Search ecosystem
- Architecture of a search engine
- Lucene-based search
- Vector search
Introduction to LLMs and Prompt Engineering
- Brief history of LLMs
- The LLM ecosystem
- Limits
- The best Prompt
Understanding the structure of a RAG system
- Architecture
- Tooling
- Gains
Five steps to creating a RAG
- Creating a Chat service
- Extracting embeddings
- Understanding the queries
- Creating a Prompt
- Behave as a bot
More information
Next sessions
- 7th June 2024
- Or on demand
Prerequisites
- Basic knowledge of HTTP and REST
- Basic knowledge of software programming
- Laptop with Java 17 minimum
- IDE and command-line terminal
Teaching method
- Presentation and demonstrations 40%
- Hands-on labs 40%
- Q&A 20%
Speakers
- Vincent Brehin
- Expert consultant in Search, Data, Solr, Elasticsearch and OpenSearch - LinkedIn
- Benjamin Dauvissat
- Expert Consultant in Search, Data, Solr, Elasticsearch and OpenSearch - LinkedIn
- blog articles
- Aline Paponaud
- Expert consultant in Search, Data, Elasticsearch and OpenSearch - LinkedIn
- Lucian Precup
- Technical director and principal consultant - LinkedIn
- blog articles
For more information, feel free to contact us.
Register