If you’ve been hearing words like AI agents, RAG, LangChain, LlamaIndex, LangGraph and wondering,
“Where do I even start learning all this?” — you’re not alone.
The good news? You don’t need a PhD to build modern AI agent apps.
DeepLearning.AI (Andrew Ng’s platform) and partners have released a whole set of short, practical courses that help you go from “I’m curious” to “I can actually build this” — step by step.
This blog is a simple guide to 24 hand-picked AI & agentic application courses.
You can use it as a roadmap: start from basics, then move into multi-agent systems, RAG, tools, browser & voice agents, and more.
Let’s dive in. 👇
Why Learn AI Agents and Agentic Workflows Now?
Traditional “chatbot only” use of LLMs is slowly becoming old news.
The future is about:
- AI agents that can plan, call tools, use APIs, browse, write code, and act on your behalf
- Multi-agent systems where several agents collaborate (researcher, writer, evaluator, coder, etc.)
- Agentic apps that can remember, adapt, and work with complex data and workflows
These courses will help you:
- Automate boring tasks
- Build smarter apps for work or business
- Stand out as a developer, tech enthusiast, or AI learner
Now, let’s look at the courses in simple language.
1–2. Multi AI Agent Systems with crewAI 🧠🤝
1. Multi AI Agent Systems with crewAI
🔗 https://www.deeplearning.ai/short-courses/multi-ai-agent-systems-with-crewai/
This course shows you how to work with crewAI, a framework where you create “crews” of AI agents that work together.
You’ll learn how to design agents with different roles (like a researcher, writer, or reviewer) and make them collaborate on a task.
2. Practical Multi AI Agents and Advanced Use Cases with CrewAI
This is like level 2 of crewAI.
Here you go beyond basics and explore advanced workflows and real-world projects — for example, automating business tasks or building complex pipelines where multiple agents coordinate.
3. Serverless Agentic Workflows with Amazon Bedrock ☁️
🔗 https://www.deeplearning.ai/short-courses/serverless-agentic-workflows-with-amazon-bedrock/
If you’re interested in cloud-native AI, this course shows how to use Amazon Bedrock to build agentic workflows without managing servers.
You’ll learn to connect tools, orchestrate agents, and deploy workflows that can scale on AWS.
4 & 7. AI Agents and Long-Term Memory with LangGraph 🔁
4. AI Agents in LangGraph
🔗 https://www.deeplearning.ai/short-courses/ai-agents-in-langgraph/
LangGraph lets you build graph-based agent workflows.
Instead of a simple question-answer flow, you design stateful, multi-step agents that can loop, branch, and handle complex logic. This course walks you through how to structure and run such agents.
7. Long-Term Agentic Memory with LangGraph
🔗 https://www.deeplearning.ai/short-courses/long-term-agentic-memory-with-langgraph/
This one focuses on memory.
You learn how your agents can remember past interactions, store important info, and use it later — like a personal AI assistant that truly knows you or a system that tracks long projects over time.
5. AI Agentic Design Patterns with AutoGen 🧩
🔗 https://www.deeplearning.ai/short-courses/ai-agentic-design-patterns-with-autogen/
AutoGen gives you a framework of patterns for building multi-LLM and multi-agent systems.
This course explains common design patterns (like helper agents, critic agents, etc.) and how to combine them to build robust, modular agentic apps.
6 & 12 & 22. Event-Driven & Agentic RAG with LlamaIndex 📄⚙️
6. Event-Driven Agentic Document Workflows with LlamaIndex
🔗 https://www.deeplearning.ai/short-courses/event-driven-agentic-document-workflows/
12. Event-Driven Agentic Document Workflows
🔗 https://www.deeplearning.ai/short-courses/event-driven-agentic-document-workflows/
(Both links point to the same core idea.)
These focus on building document workflows that are event-driven — for example:
- A new file gets uploaded → an AI agent processes and summarises it
- A contract changes → an agent highlights differences
- A folder is updated → an agent refreshes your knowledge base
You use LlamaIndex to organise and query documents smartly with agents.
22. Building Agentic RAG with LlamaIndex
🔗 https://www.deeplearning.ai/short-courses/building-agentic-rag-with-llamaindex/
Here you go deeper into RAG (Retrieval-Augmented Generation) + agents.
Instead of a simple RAG chatbot, you build agentic RAG where agents decide how to search, what to fetch, and how to combine info for better answers.
8, 14 & 23. Coding Agents & Claude Code 💻
8. Build Apps with Windsurf’s AI Coding Agents
🔗 https://www.deeplearning.ai/short-courses/build-apps-with-windsurfs-ai-coding-agents/
This is perfect if you’re a developer (or learning to code) and want AI to help you build apps faster.
You learn how Windsurf’s AI coding agents can generate, refactor, and debug code while you stay in control.
14. Building Code Agents with Hugging Face
🔗 https://www.deeplearning.ai/short-courses/building-code-agents-with-hugging-face-smolagents/
This course uses Hugging Face smolagents to build lightweight but powerful code-focused agents.
You’ll see how agents can run tools, interact with code, and help you automate coding tasks.
23. Claude Code (Anthropic)
Here you learn Claude Code, Anthropic’s highly agentic coding assistant.
You’ll see how to use Claude for multi-file edits, refactoring large codebases, and building an agent that understands your entire project context.
9, 19 & 24. LangChain & Haystack – Building LLM Applications 🧱
9. Building AI Applications with Haystack
🔗 https://www.deeplearning.ai/short-courses/building-ai-applications-with-haystack/
Haystack is an open-source framework for search, RAG, and question answering.
This course teaches you how to build real-world apps that answer questions from your own documents using LLMs.
19. LangChain for LLM Application Development
🔗 https://www.deeplearning.ai/short-courses/langchain-for-llm-application-development/
If you want a solid base in LangChain, start here.
You learn how to chain prompts, tools, memory and models to build practical LLM applications.
24. Functions, Tools and Agents with LangChain
🔗 https://www.deeplearning.ai/short-courses/functions-tools-agents-langchain/
This is the next step into more advanced LangChain usage.
You’ll see how to use function-calling, tools, and agents so your AI can browse, call APIs, and perform actions instead of just chatting.
10, 11, 18 & 21. Accuracy, Evaluation & RAG 📊
10. Improving the Accuracy of LLM Applications
🔗 https://www.deeplearning.ai/short-courses/improving-accuracy-of-llm-applications/
AI that “sounds confident but is wrong” is dangerous.
This course shows you methods to reduce hallucinations, improve grounding, and design higher-quality LLM apps.
11. Evaluating AI Agents
🔗 https://www.deeplearning.ai/short-courses/evaluating-ai-agents/
Once you build agents, you must measure how well they perform.
This course teaches you to evaluate agents using metrics, scenarios, and tests, so you know what to improve.
18. Function-Calling and Data Extraction with LLMs
🔗 https://www.deeplearning.ai/short-courses/function-calling-and-data-extraction-with-llms/
Here you learn how to make LLMs call tools and extract structured data from messy text.
Very useful for building agents that interact with APIs, databases, and external systems.
21. Building and Evaluating Advanced RAG Applications
🔗 https://www.deeplearning.ai/short-courses/building-evaluating-advanced-rag/
Go beyond basic RAG.
You’ll explore advanced retrieval, hybrid search, evaluation of RAG quality, and methods to truly trust your knowledge-based AI systems.
13, 15 & 20. Browser, Voice & Database Agents 🌐🎙️🗄️
13. Building AI Browser Agents
🔗 https://www.deeplearning.ai/short-courses/building-ai-browser-agents/
This course is about agents that can use the browser like a human: click, scroll, read pages, extract info.
Great for building research assistants, automation bots, and data-gathering tools.
15. Building AI Voice Agents for Production
🔗 https://www.deeplearning.ai/short-courses/building-ai-voice-agents-for-production/
Here you learn how to build voice-based AI agents that can be used in real products — think customer support bots, phone assistants, or voice-enabled tools.
20. Building Your Own Database Agent
🔗 https://www.deeplearning.ai/short-courses/building-your-own-database-agent/
This one focuses on agents that talk to your database.
You’ll learn how to let users ask questions in natural language while the agent translates them into queries, retrieves data, and returns clear answers.
16 & 17. DsPy and MCP – Advanced Agentic Apps 🧪
16. DsPy: Build and Optimise Agentic Apps
🔗 https://www.deeplearning.ai/short-courses/dspy-build-optimize-agentic-apps/
DsPy is about optimising prompts and workflows programmatically.
You learn how to treat prompt design as something you can tune and improve with code, making your agents smarter over time.
17. MCP: Build Rich-Context AI Apps with Anthropic
🔗 https://www.deeplearning.ai/short-courses/mcp-build-rich-context-ai-apps-with-anthropic/
MCP (Model Context Protocol) is a new way to give AI rich context from multiple tools and sources.
This course shows how to connect different systems so your agents have a fuller “view of the world” when answering or acting.
Unlock the Future: Your Ultimate AI Agents Learning Toolkit is Here
Introducing the AI Agents Learning Toolkit—a meticulously curated bundle designed by industry pioneers to transform you from novice to builder. Forget fragmented tutorials and outdated theory. This is your all-in-one launchpad into the frontier of autonomous AI.
How to Use This List as a Learning Roadmap 🗺️
If you’re just starting:
- Begin with:
- LangChain for LLM Application Development (19)
- Building AI Applications with Haystack (9)
- Then move into agents & tools:
- Functions, Tools and Agents with LangChain (24)
- Function-Calling and Data Extraction (18)
- Next, explore agent frameworks:
- Multi AI Agent Systems with crewAI (1)
- AI Agents in LangGraph (4)
- AI Agentic Design Patterns with AutoGen (5)
- Finally, go deeper into RAG, memory & evaluation:
- Building Agentic RAG with LlamaIndex (22)
- Building and Evaluating Advanced RAG (21)
- Evaluating AI Agents (11)
Pick what matches your current level and goals. You don’t have to do everything at once.
Final Thoughts: 2025 Is the Year of AI Agents – Will You Miss It?
AI is moving from “chat with a bot” to “give a task to an intelligent agent and let it handle the workflow.”
These 24 short courses give you a clear, structured way to:
- Understand AI agents
- Build real applications
- Improve accuracy, reliability, and performance
- Level up your skills for the future of work
You can bookmark this post as your AI agent learning checklist.
Even if you finish just a few of these courses, you’ll be miles ahead of most people still stuck at basic prompt typing.
If you want, I can also help you turn this list into a personal learning plan (beginner / intermediate / advanced) based on how much time you can spend each week.
Subscribe to our channels at alt4.in or at Knowlab
Have a take? Say it on Reddit. We’d love your perspective—comment or views.
