Skip to Content
View all events

Fundamentals of Large Language Models Bootcamp

Published by O'Reilly Media, Inc.

Beginner content levelBeginner

Fundamentals of Large Language Models Bootcamp: A practical guide to modern AI Systems

Course outcomes:

  • Demonstrate how to prompt (and interact) with Large Language Models.
  • Recognize when Large Language Models can and cannot be used.
  • Determine when and how to employ embeddings from Large Language Models.

Chat-GPT has captivated the public’s interest like few other AI-related concepts. Suddenly there has been a surge in demand for Large Language Models. The research community has been able to build upon several significant ideas from GPT-3 such as self-supervised learning, the scaling laws (Chinchilla paper), and the significance of reinforcement learning (DeepMind Sparrow) amongst others.

As part of this training, we will discuss these concepts along with major LLM contributions since GPT-3. We will also do hands-on exercises using a Large Language Model to perform different tasks such as generating LinkedIn posts, summarizing news items, and embeddings, among others.

What you’ll learn and how you can apply it

  • How we got to where we are from ChatGPT.
  • How to compare and determine which LLM is right for your organization.
  • How you can use embeddings from Large Language models for search-related tasks.

This live event is for you because...

  • You’re a Machine Learning Engineer/Data Scientist
  • You work with LLMs and want to understand what is happening behind the scenes

Prerequisites

  • Accounts on openai.com and cohere.ai (necessary to participate in exercises)
  • Basic understanding of machine learning (ML model, parameters)

Recommended follow-up:

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Day 1

Foundations (60 minutes)

  • Presentation: Fundamentals - Parameters, tokens, context length, prompting
  • Presentation: Transformer Architecture
  • Presentation: Dense vs. Mixture of Expert and current trends
  • Q&A
  • Break

Training LLMs (60 minutes)

  • Presentation: Scaling Laws
  • Presentation: How LLMs are trained
  • Exercise: Benchmarks
  • Q&A
  • Break

Customizing Models for Real-World Use (60 minutes)

  • Presentation: Fine-tuning
  • Exercise: Fine-tuning an LLM with a small dataset
  • Presentation: Quantization
  • Q&A
  • Break

Deploying LLMs in production (60 minutes)

  • Presentation: Open and closed models
  • Presentation: Deploying models in production
  • Q&A
  • Break

Day 2

LLM embeddings lab (60 minutes)

  • Presentation: LLM embeddings
  • Demo: Embeddings
  • Demo: Multilingual embeddings
  • Exercise: Creating embeddings
  • Q&A
  • Break

Beyond chatbots (60 minutes)

  • Presentation: Agents
  • Presentation: Tools and standards (MCP)
  • Exercise: Model Context Protocol
  • Q&A
  • Break

Working with LLMs (60 minutes)

  • Presentation: Context Engineering for LLMs
  • Demo: LLM forgetting and fixing it
  • Q&A
  • Break

Evals (60 minutes)

  • Presentation: What are evals?
  • Presentation: Key considerations when working with evals
  • Demo: How to work with evals - an example from one of my projects.
  • Q&A

Your Instructor

  • Jonathan Fernandes

    Jonathan Fernandes works with large language models every day, and he did so long before ChatGPT came on the scene. He focuses primarily on LLMs in production to solve business problems across a variety of domains including finance, health, manufacturing, legal, and education. Previously, he worked for LLM-provider startups including Cohere. Jonathan has taught over 250,000 people about LLMs and how to use them.

Skill covered

Large Language Models (LLMs)