AI Engineering 2026: ML, Deep Learning, LLMs, RAG & Agents

About this course

Become an AI engineer, from your first line of Python to fine-tuning your own LLM.This course takes you through the full AI engineering stack in one structured path: Python, machine learning, deep learning, generative AI, AI agents, RAG and LLM fine-tuning. It was recorded from live, hands-on classes, so you see real code, real tools and real problems worked through step by step.

What you'll cover

Foundations: the AI landscape, your AI workspace (Google Colab, GitHub, Hugging Face, VS Code) and Python for AI the 80/20 way.

Machine learning: data preprocessing with pandas and scikit-learn, exploratory data analysis, regression, classification, decision trees, random forests, XGBoost, SVMs and k-means, plus overfitting, loss functions and evaluation metrics.

Deep learning: neural networks from the perceptron up, backpropagation and gradient descent, training in PyTorch, CNNs for computer vision, transfer learning, RNNs and NLP.Generative AI: tokens, embeddings, attention and transformers, context windows, quantization, mixture-of-experts, and choosing between open and proprietary models.

AI agents: MCP, agent frameworks, multi-agent workflows and building a deep research agent with LangGraph.

RAG and AI safety: embeddings, vector databases, agentic RAG, guardrails and the OWASP Top 10 for LLMs.

Build and fine-tune LLMs: train a small story-writing LLM from scratch, then fine-tune an open-weight model with LoRA/QLoRA and Unsloth and publish it to Hugging Face.

Coding with AI agents: use IDE and terminal coding agents to build and ship ML apps faster.

Your AI career: roles, portfolio, resume and what interviews really test.

Read the full description on Udemy on Udemy (opens in a new tab)

Prices, discounts and availability are set by Udemy. We may earn a commission when you purchase through links on this site.