15 lesson learning series

Artificial intelligence has acquired a vocabulary of its own: machine learning, neural networks, tokens, embeddings, transformers, pre-training, RLHF, RAG, and fine-tuning. The words often arrive faster than the explanations.

This series is a learning path, not a glossary. Each lesson gives you the mental model needed for the next one.

15 lessons 6 modules Beginner starting level

Start with Lesson 1 →

You do not need a machine-learning background to begin. We start with the relationship between AI, machine learning, and deep learning, then follow the path from numerical representations and learning methods into neural networks, transformers, training, alignment, retrieval, and model adaptation.

The goal is not to make every topic sound simple. It is to make each topic understandable without making it inaccurate.

What you will understand by the end

By the final lesson, you should be able to explain:

  • how AI, machine learning, and deep learning relate;
  • how text, images, and audio become numerical representations;
  • the differences among supervised, unsupervised, self-supervised, and reinforcement learning;
  • how self-supervised objectives create training signals from raw data;
  • what neural-network weights actually do;
  • what a training corpus contributes and what it does not;
  • how transformers use attention, including the intuition behind queries, keys, and values;
  • how forward passes, loss, backpropagation, and optimization fit together;
  • why large training runs consume so much compute;
  • how post-training turns a base model into a useful assistant;
  • how RLHF, RLAIF, and DPO differ within preference-based post-training;
  • why RAG can ground answers without eliminating hallucinations;
  • when prompting, retrieval, and fine-tuning are the right tools.

How to use the series

If this is your first serious introduction to AI, read the lessons in order. They are intentionally focused, and the sequence builds from first principles.

If you already work with AI systems, use the modules as a reference. The lessons remain independent enough to revisit when you need a clear explanation of a particular concept.

Module 1

The Foundations

  1. 1AI vs. ML vs. Deep Learning
  2. 2How Machines Represent Data as Numbers

Module 2

How Machines Learn

  1. 3Supervised Learning
  2. 4Unsupervised Learning
  3. 5Self-Supervised Learning
  4. 6Reinforcement Learning

Module 3

Deep Learning and Large Language Models

  1. 7Artificial Neural Networks
  2. 8What Is a Corpus?
  3. 9How Transformers Predict the Next Token

Module 4

Inside the Training Loop

  1. 10The Neural Network Training Loop
  2. 11Why LLM Pre-Training Uses So Many Tokens

Module 5

Post-Training and Alignment

  1. 12Pre-Training vs. Post-Training
  2. 13RLHF, RLAIF, and DPO

Module 6

Applying Large Language Models

  1. 14RAG Explained
  2. 15Fine-Tuning vs. Prompt Engineering

After the fundamentals

Continue with How AI Systems Get Work Done, an eight-lesson series explaining tools, agent loops, harnesses, workflow graphs, and multi-agent systems through one practical example.

This series explains the foundations. The rest of jacobpallattu.com explores what happens when those models become part of real software systems: retrieval, tools, agents, governance, architecture, evaluation, and experimental model work.

Continue into the experiments

PallattuLM connects these concepts to a model being built and trained in public view. The architecture and lab sections then move from model fundamentals into real AI system design.

Explore PallattuLM →   Browse AI architecture and systems →