Welcome to
AI Bootcamp
Course Syllabus and Content
Introduction to AI applications
8 Lessons
This module introduces foundational concepts and practical workflows for working with Large Language Models (LLMs). Topics include terminology (e.g., ChatGPT vs. LLM, inference phases, training stages, and model compression techniques), the LLM ecosystem (vector databases, inference APIs, and fine-tuning libraries), and the model lifecycle. Participants will build a simple LLM-based system from scratch, starting with “Hello World” inference using Hugging Face, and deploy an LLM API using Modal for serverless deployment.
- 01Introduction to AI applicationsSneak Peek
- 02Intro to AI modalitiesSneak Peek
- 03Examples of AI applicationsSneak Peek
- 04TechnologiesSneak Peek
- 05How to brainstorm AI applicationsSneak Peek
- 06Vertical applications and their example technologiesSneak Peek
- 07Indie hacker examplesSneak Peek
- 08Venture funded LLM examplesSneak Peek
Building a Shakespearean Language Model
7 Lessons
Building a Shakespearean Language Model
- 01What is a tokenizer? What makes a tokenizer "good"?Sneak Peek
- 02Build a baseline 'word-based' tokenizerSneak Peek
- 03Build a byte-pair encodingSneak Peek
- 04Compare with state-of-the-art tokenizers (e.g., Llama’s)Sneak Peek
- 05What is an embedding? What makes an embedding "good"?Sneak Peek
- 06Understand the semantic meaning of word embeddingsSneak Peek
- 07Train a word embedding on Shakespearean tokensSneak Peek
Building an n-gram language model
5 Lessons
Building an n-gram language model
- 01What is an n-gram?Sneak Peek
- 02Train a 2-gram modelSneak Peek
- 03Predict using 2-gram modelSneak Peek
- 04Train an n-gram modelSneak Peek
- 05Predict using n-gram model, in an autoregressive mannerSneak Peek
Building self-attention
2 Lessons
Building self-attention
- 01A minimal version of self-attentionSneak Peek
- 02Build a batched version of self-attentionSneak Peek
Building the feed-forward neural network
3 Lessons
Building the feed-forward neural network
- 01A minimal version of the feed-forward neural network (e.g., MLP)Sneak Peek
- 02Build a batched version of the MLPSneak Peek
- 03“Train” on sanity check dataSneak Peek
Assembling the transformer-based language model
3 Lessons
Assembling the transformer-based language model
- 01Add the remaining components (Skip connections, norms, positional encodings)Sneak Peek
- 02Use black box optimizer to pretrainSneak Peek
- 03Predict using transformer-based language model, autoregressivelySneak Peek
Evaluating and deploying a transformer-based language model
4 Lessons
Evaluating and deploying a transformer-based language model
- 01Run unsupervised evaluation using perplexitySneak Peek
- 02Export the model using torchscriptSneak Peek
- 03Plug in our custom model into our LLM APISneak Peek
- 04Open-sourced pre-trained models include Llama, Phi-3, Mixtral etc.Sneak Peek
Datasets
3 Lessons
Datasets
- 01Pre-curated pretraining datasetsSneak Peek
- 02Common tools for data curation (mech turks, vis tools)Sneak Peek
- 03how to add guardrailsSneak Peek
Low-Rank Adapters for Instruction Tuning
4 Lessons
Low-Rank Adapters for Instruction Tuning
- 01Build low-rank adapters for Shakespearean modelSneak Peek
- 02Download instruction tuning datasetSneak Peek
- 03Run low-rank adapter fine-tuningSneak Peek
- 04Open-sourced instruction-tuned models include DBRX, Pythia, Cerebrus etc.Sneak Peek
Retrieval-Augmented Generation (RAG)
5 Lessons
Retrieval-Augmented Generation (RAG)
- 01What is RAG?Sneak Peek
- 02Implement RAGSneak Peek
- 03Apply RAG to Shakespearean modelSneak Peek
- 04Apply RAG to pre-trained Large Language ModelsSneak Peek
- 05Integrate RAG into our APISneak Peek
The Future of Large Language Models
5 Lessons
The Future of Large Language Models
- 01How to use LlamaIndex to chain tools togetherSneak Peek
- 02The data wallSneak Peek
- 03The ultimate test of time (Use fundamentals to understand and use time to filter)Sneak Peek
- 04How to keep to dateSneak Peek
- 05Multi modal large language modelsSneak Peek
Machine learning operations
4 Lessons
Machine learning operations
- 01project scopingSneak Peek
- 02data needsSneak Peek
- 03modeling strategiesSneak Peek
- 04deployment requirementsSneak Peek
Agents
2 Lessons
Agents
- 01What are agents?Sneak Peek
- 02Design patternsSneak Peek