Institution
Udemy
Course snapshot
- Agentic AI
- Generative AI
- Reinforcement Learning
- Q-Learning
- Deep Q-Learning (DQN)
- Convolutional Deep Q-Learning
- A3C (Asynchronous Advantage Actor-Critic)
- PPO (Proximal Policy Optimization)
- SAC (Soft Actor-Critic)
- Transformers
- Large Language Models (LLMs)
- LoRA & QLoRA fine-tuning
- Text classification & extraction
- Prompt design for LLMs
- Knowledge augmentation for LLMs
This 2025 edition of Artificial Intelligence A-Z blends agentic AI, generative AI, and reinforcement learning into a single practical track. You start with hands-on prompting basics, how to shape factual answers, summaries, extractions, classifications, conversations, code, and reasoning, and learn to evaluate responses with objective and subjective metrics and parameter tuning. From there you build a suite of reinforcement learning systems: classic Q Learning, Deep Q Learning, and a convolutional DQN for games like Pac Man and control tasks like a lunar lander. You then move to policy gradient families including A3C, PPO, and SAC for more robust decision making, including self driving simulations. On the generative side, you work with transformers and large language models, applying LoRA and QLoRA to fine tune efficiently and using knowledge augmentation to specialize an LLM, for example a medical chatbot. The course code is built in Google Colab to avoid setup pain, with downloadable .py and .ipynb templates so you can remix quickly. Along the way, intuition first explanations keep the math approachable while still giving you a correct mental model of how each algorithm learns. Extras include DDPG, world models, and evolution strategies and genetic algorithms, plus a bonus mini course on generative AI and cloud. By the end you have multiple working AIs, agentic LLM apps and RL agents, plus a toolkit for evaluating prompts, tuning models, and turning ideas into runnable systems.
Pricing and availability can change. Always check the provider page.
Build practical AIs—agents, LLM apps, and reinforcement learners—by coding them end-to-end.— Course team on Udemy
Syllabus
-
Course Overview
-
Introduction to Prompting & Real-World Examples
-
Creating Effective Prompts (factual, summarization, extraction, classification, conversation, code, reasoning)
-
Evaluating Prompt Performance (objective/subjective metrics, parameter tuning)
-
Advanced Prompt Techniques (zero/few-shot, chain-of-thought, least-to-most, generated knowledge)
-
RL Foundations: Q-Learning, Deep Q-Learning, Conv DQN
-
RL Advanced: A3C, PPO, SAC
-
LLMs & GenAI: Transformers, LoRA/QLoRA, fine-tuning with knowledge augmentation
-
Projects: Chatbot with a foundation model; warehouse optimization; lunar lander; Pac-Man; Kung Fu; self-driving car; medical chatbot
-
Extras: DDPG, World Models, Evolution Strategies & Genetic Algorithms
Instructors & Institution
Admission & cost
30-day money-back guarantee on Udemy. Course also available via Udemy Personal Plan subscription.
You might also like
Data Science MicroMasters (UC San DiegoX)
Open →
Computer Science for Artificial Intelligence (Professional Certificate)
Open →
Making AI Work: Machine Intelligence for Business and Society
Open →
Artificial Intelligence: Getting Started
Open →
LLM Engineering: Master AI, Large Language Models & Agents
Open →
AI Ethics, Law, and Policy
Open →Recent updates
Last updated: