Course Outline
Day 1 – Generative AI & LLM Fundamentals
Introduction to generative AI and practical use cases for LLMs
Exploring transformer-based architectures (GPT, LLaMA, T5, etc.)
Understanding tokens, tokenization, and embeddings
Leveraging pre-trained models through APIs (OpenAI, Claude)
Utilizing pre-trained models via Hugging Face
Core prompting techniques: zero-shot and few-shot approaches
Hands-on exercise: prompt engineering in a Python notebook
Developing a straightforward LLM-powered application (CLI or web-based)
Addressing practical constraints: token limits, rate limitations, and basic reliability strategies
Day 2 – RAG and Vector Search
The rationale behind RAG: merging LLMs with proprietary data
RAG architecture breakdown: ingestion, indexing, retrieval, and generation
Preparing documents through effective chunking for optimal retrieval
Generating text embeddings using APIs or Hugging Face models
Overview of vector databases (e.g., Chroma, Pinecone)
Hands-on exercise: constructing a basic semantic search script
Hands-on project: building a document Q&A system with RAG
Scaling ingestion and embedding pipelines for larger datasets
Evaluating design trade-offs in RAG: chunk size, top-k selection, cost versus quality balance
Day 3 – Workflows, Agents, and Production
Defining AI agents and identifying suitable application scenarios
Introduction to LangGraph and graph-based LLM workflows
Hands-on exercise: creating a simple LangGraph workflow with integrated tools
Enhancing workflows with memory capabilities and multi-step reasoning
Integrating RAG with agentic frameworks for advanced functionality
Monitoring and evaluating performance of LLM and RAG systems
Deployment strategies for LLM applications (APIs, containers, services)
Optimizing costs and system performance effectively
Implementing basic safety measures, guardrails, and responsible usage practices
Capstone mini-project: delivering an end-to-end RAG/agent application demonstration
Requirements
Solid Python programming skills and familiarity with APIs are required.
Target Audience:
This course is designed for organizations aiming to transition from experimentation to fully operational LLM-powered solutions. It caters to software, backend, and full-stack engineers integrating LLMs into their products; data and machine learning engineers focusing on RAG, embeddings, and vector search; solution and enterprise architects crafting LLM-based architectures; as well as technical product owners and engineering leaders tasked with assessing AI use cases, costs, and risks.
Testimonials (2)
use of proper and effective prompt
Marses Pacaldo
Course - Generative AI and Prompt Engineering for Corporate Professionals
The interactive style, the exercises
