AI Study

The engineer's path into AI. From machine learning fundamentals to production LLM systems — practical, structured, and built for people who already ship software.

Start learning Interview prep

You already know how to build software. This guide teaches you the other half of modern AI engineering — how models work, how to wire them into real systems, and how to make them fast, cheap, and reliable in production. No PhD required. Every topic is written for developers, with code, diagrams, and clear trade-offs.

Learning paths

Pick a track

Foundations

New to ML

Start with how models learn, then deep learning and transformers. The mental models everything else builds on.

Applied

LLM App Developer

Skip straight to building. LLM engineering, prompt design, RAG, agents, multimodal, and AI system design.

Production

Platform & MLOps

Serving, GPUs, scaling, cost control, monitoring, security, and the operational side of AI systems.

Career

Interviewing

Concept refreshers, LLM system-design drills, and coding questions for AI engineering roles.

The full curriculum

All 17 topics

AI Fundamentals

What AI actually is, how the field is organized, and how models learn.

Machine Learning

Learning paradigms, evaluation, features, and the core algorithms.

Deep Learning

Neural networks, backprop, training dynamics, and key architectures.

LLM Engineering

How LLMs work, context windows, decoding, and adapting models.

AI System Design

Designing LLM applications for cost, latency, and reliability.

Vector Databases

Embeddings, similarity search, indexing, and choosing a store.

RAG Architecture

Retrieval pipelines, chunking, advanced RAG, and evaluation.

AI Agents

Agent loops, tool use, memory, and multi-agent systems.

Prompt Engineering

Core techniques, structured output, and reliable prompt patterns.

Multimodal AI

Vision-language models, image generation, speech, and voice agents.

AI Infrastructure

Model serving, GPUs, inference scaling, and cost optimization.

MLOps

The ML lifecycle, deployment, monitoring, and LLMOps.

AI Safety & Security

Prompt injection, data privacy, and responsible AI for builders.

AI Coding Tools

The landscape, working effectively, and agentic coding workflows.

Open-Source AI

Open models, frameworks, and running models on your own hardware.

Interview Prep

Concepts, LLM system design, coding, and behavioral rounds.

Architecture Patterns

Battle-tested RAG, agent, and production patterns with case studies.

New to the vocabulary? Keep the Glossary open as you read — it defines every term the guide uses.

Why this guide

Built for engineers, not researchers

Practical first

Every concept is tied to something you’d actually build or debug. Theory shows up only where it changes a decision.

Code & diagrams

Runnable snippets, request/response shapes, and architecture diagrams instead of dense notation.

Trade-offs explicit

”When to use what” guidance — cost, latency, accuracy, and complexity stated up front, not buried.

Beginner to advanced

Each section ramps from first principles to production concerns, so you can enter at your level.

Ready to start?

Brand new to AI

Begin with AI Fundamentals to build the core mental models, then move through Machine Learning and Deep Learning.

Building an LLM app

Jump to LLM Engineering, then RAG, AI Agents, and AI System Design.

Shipping to production

Head to AI Infrastructure and MLOps for serving, scaling, monitoring, and cost control.

Prepping for interviews

Use Interview Prep alongside Architecture Patterns to rehearse real design questions.