Workshop of Basic LLM app

• Prompt engineering
  • Zero shot and CoT
    • Zero shot
    • CoT
    • Large Language Models are Zero-Shot Reasoners
  • React
• Basic of LLM
  • Transformer process
    • Tokenizer
    • Embedding
    • Similarity
  • Text-to-SQL
• Problem and solution ?
• Caching
  • Context/prompt
    • OpenAI
    • Anthropic
  • Response/semantic
• Function calling
  • Working with OpenAI
  • React with langchain
• React
  • Demo with React
• MCP = Model Context Protocol
  • Demo
• RAG (Retrieval-Augmented Generation)
  • RAG workshop
    • Data source = Website
    • Vector database = Chroma
  • Processing
    • Chunking or splitting
    • Embedding
    • Store data in to Vector database
    • Similarity or Relevance search
    • Re-ranking
    • Evaluate
• CAG (Cache-Augmented Generation)