Last updated: March 03, 2026
By M. Ali Yousuf
[Disclaimer: The code on this page is NOT written by me and comes from the original author/website mentioned in each Google Colab notebook. I may have made modifications, when needed, to make them run from within Google Colab, and some extra code/text added to explain the code.]
The main difference between AI agents and agentic AI is their scope, autonomy, and adaptability. AI agents are designed to perform specific tasks within defined boundaries, operating independently but only within their programmed rules. In contrast, agentic AI refers to systems capable of orchestrating multiple agents to autonomously handle complex, multi-step workflows, adapt to new circumstances, and pursue goals proactively with advanced reasoning and learning capabilities.
According to Andrew Ng, Agentic AI refers to AI systems that move beyond single-prompt, text-in/text-out interactions to autonomously achieve complex goals through iterative, multi-step workflows. These systems utilize tools, reasoning, and self-reflection to improve performance, acting as agents that collaborate or work independently with minimal human guidance.
Typically single entities designed for focused, well-defined tasks (e.g., chatbots, scheduling assistants).
Operate with rule-based logic or machine learning algorithms, making decisions within strict parameters and boundaries set by their programming.
Reactive: act in response to specific inputs or triggers rather than proactively identifying new opportunities or solutions.
Improvement usually relies on human intervention (model retraining or logic updates) rather than independent learning.
Represents a broader paradigm where multiple agents may collaborate or be orchestrated to achieve complex, dynamic goals.
Possesses true agency—the capability to set its own goals, strategize, and adapt based on real-time conditions and outcomes.
Proactive: not only responds to requests but also takes initiative, plans multi-step actions, and adapts strategies on the fly.
Self-improving and learning continuously from experience, capable of redefining boundaries and developing new solutions as needed.
MCP stands for Model Context Protocol, an open standard developed in late 2024 for connecting AI systems - such as large language models (LLMs) and agents - to external tools, data sources, and services. MCP enables AI agents to access, read, and act on information outside their training data, such as files, business applications, or specialized databases, using a standardized protocol similar to an API but specifically designed for AI interoperability.
MCP allows agentic AI systems and AI agents to move beyond simple retrieval or chat, enabling complex, real-world automation (e.g., autonomous agents booking meetings, aggregating business insights, or acting in business management tools). Tools let LLMs take actions through your server. Unlike resources, tools are expected to perform computation and have side effects.:
By providing structured, secure access to external context and tools, MCP serves as a foundational protocol for next-generation agentic AI applications.
The Complete Guide to Model Context Protocol By Bala Priya C on October 29, 2025 in Artificial Intelligence, https://machinelearningmastery.com/the-complete-guide-to-model-context-protocol/
Book: Christoffer Noring, "Learn Model Context Protocol with Python," First Edition
Corresponding github page: https://github.com/PacktPublishing/Learn-Model-Context-Protocol-with-Python/tree/main?tab=readme-ov-file
Now that you understand the difference between AI Agents, Agentic AI and the role of MCP, let's look at these things in more detail. Below is a collection of labs to give you some hands-on experience. Please note that I am NOT the author of these examples - I have just collected them here with some additional notes in the notebooks. You will find links to the original versions within the notebooks.
A note on Google Colab notebooks
When you will click on any file below, it will open and you will be able to make changes. However, you will NOT be able to save those changes as these are my files and you are only a viewer. To get full access, make a 'copy' by clicking on File -> 'Save a Copy in Drive'. That way, you will have a copy of the file in YOUR google drive, under the folder 'Colab Notebooks' (generally a yellow colored folder symbol). You can edit that version as much as you want and it will be saved on your drive. In case you mess up, go back to the original link to the file in my Google Drive and make another fresh copy!
(Recommended for DATA 606 students) Andrew Ng's course 'Agentic AI' (free version good enough), https://www.deeplearning.ai/courses/agentic-ai/
Andrew Ng, STUNNING AI Architecture Revealed, https://www.youtube.com/watch?v=wM5837pVh1g
Andrew NG, Agentic Document Extraction, https://youtu.be/Yrj3xqh3k6Y?si=sbxLi4_rGBiFIBWl
IBM Technology - What are AI Agents? https://www.youtube.com/watch?v=F8NKVhkZZWI&t=625s
Crash Course on Agent Engineering - Learn about Agentic Workflows, Design Patterns and Good Practices for AI Agent Engineering, by Pascal Biese, https://www.llmwatch.com/p/ai-agents-crash-course-on-agent-engineering
LlamaIndex - AgentWorkflow Basic Introduction, https://docs.llamaindex.ai/en/stable/examples/agent/agent_workflow_basic/
Watch the YouTube series "Agentic AI Learning Series", https://youtube.com/playlist?list=PLuz5G1-Qs503EGrPNu4IP6prmNIYJ2rP7&si=u-nsfMH2iHAOvM1E
Anthropic's Building effective agents, https://www.anthropic.com/research/building-effective-agents
Strands, https://strandsagents.com/ allows you to build production-ready, multi-agent AI systems in a few lines of code
It allows you to use OpenAI, Ollama, etc.
Here is a Google Colab example using OpenAI (you need OpenAI API key). https://colab.research.google.com/drive/14WXKE5PetcUbvR6lkF6LgliC4aTseH5e?usp=sharing.
Here is a similar example using Gemini (you need Gemini API key which is free), https://colab.research.google.com/drive/1Wwv3iNt005prUhimJqOPy4cOXJPetbNL?usp=sharing
Gain control with LangGraph to design agents that reliably handle complex tasks, https://www.langchain.com/langgraph
Pydantic AI is a Python agent framework designed to help you quickly, confidently, and painlessly build production grade applications and workflows with Generative AI. https://ai.pydantic.dev/
The OpenAI Agents SDK is a lightweight yet powerful framework for building multi-agent workflows. It is provider-agnostic, supporting the OpenAI Responses and Chat Completions APIs, as well as 100+ other LLMs. https://github.com/openai/openai-agents-python
Using OpenAI Swarm in Python: A Step-by-Step Guide, https://www.launchfa.st/blog/using-openai-swarm
The leading Multi-Agent Platform - Streamline workflows across industries with powerful AI agents. Build and deploy automated workflows using any LLM and cloud platform.
A general-purpose 'Crew AI' based system to research a topic: https://colab.research.google.com/drive/1V1jgshUdzaqSpsxYxHNHEmhoXBai58rF?usp=sharing [Requires OpenAI API Key]
A multi agent system using CrewAI tools https://colab.research.google.com/drive/1xkdkOfnhtdRUbRli3Rno07ZBRJKpP6VZ?usp=sharing
Agentic Weather Data, https://colab.research.google.com/drive/1E34k8oBm9JyEvP8Xsum5eqUe4qRAEUEx?usp=sharing
A Dummy AI Agent to understand how they work [free RAPID_API_KEY required], https://colab.research.google.com/drive/1FlQ1nbL9G7Y86QRw7XQ0dFWk-P97FiVG?usp=sharing
The full version of the above code that gets the real URL access time,
Creating An AI Agent-Based System with LangGraph: A Beginner’s Guide, [not a Google Colab notebook] https://www.marktechpost.com/2025/01/29/creating-an-ai-agent-based-system-with-langgraph-a-beginners-guide/
Creating an AI Agent-Based System with LangGraph: Putting a Human in the Loop, https://www.marktechpost.com/2025/02/04/creating-an-ai-agent-based-system-with-langgraph-putting-a-human-in-the-loop/
A course that teaches you the same: Multi AI Agent Systems with crewAI, https://www.deeplearning.ai/short-courses/multi-ai-agent-systems-with-crewai/
Secure, AI-native workflow automation. The world's most popular workflow automation platform for technical teams. https://n8n.io/
It is free for 14 days after which NOT free. Starter is $20 per month.
You can download and run it via Docker and that is FREE! See instructions and further links here: https://colab.research.google.com/drive/1QZjvMdXvTle7BAjJQ4qwPqQ_KKWQHAmc?usp=sharing
Here is a video introduction by Tina Huang: https://www.youtube.com/live/NUPjbWsSe7s?si=ITsFVY3xC9rNbu9_
Langflow is a low-code tool for developers that makes it easier to build powerful AI agents and workflows that can use any API, model, or database.
Check their website: https://www.langflow.org/
Find a tutorial here: https://www.youtube.com/watch?v=VbqfZjT_PLE
There is a DataStax version of it too: https://www.datastax.com/lp/advanced-agentic-pipelines-with-langflow-a
AutoGen is a framework for building AI agents and applications
I still have to understand and use it in some way. https://github.com/pixeltable/pixeltable
A memory-based, continual-learning framework that helps LLM agents improve from experience without updating model weights. The paper https://arxiv.org/abs/2508.16153
Strands. Strands Agents is an open-source SDK that was initially released by AWS, as described in the AWS Open Source Blog. Strands Agents is designed for building autonomous AI agents with a model-first approach It provides a flexible, extensible framework designed to work seamlessly with AWS services while remaining open to integration with third-party components. Strands Agents is ideal for building fully autonomous solutions. https://aws.amazon.com/blogs/machine-learning/strands-agents-sdk-a-technical-deep-dive-into-agent-architectures-and-observability/
Bedrock. Amazon Bedrock powers generative AI for more than 100,000 organizations worldwide—from startups to global enterprises across every industry. It provides the proven infrastructure and comprehensive capabilities to confidently build applications and agents that work in production with the flexibility, enterprise security, and proven scalability you need to innovate boldly and deliver AI that drives real business impact. https://aws.amazon.com/bedrock/
Bedrock AgenticCore - Set up permissions for AgentCore Gateway, https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-prerequisites-permissions.html
During the graduate level coures on Agentic AI, we'll use the book: "Building Agentic AI Systems: Create intelligent, autonomous AI agents that can reason, plan, and adapt," by A Biswas and W Talukdar.
https://www.packtpub.com/en-us/product/building-agentic-ai-systems-9781803238753, or cheaper at Amazon: https://www.amazon.com/gp/cart/view.html?ref_=nav_cart
The corresponding code repository can be found at https://github.com/PacktPublishing/Building-Agentic-AI-Systems
Tina Huang's collection: Building AI Agents, https://docs.google.com/document/d/1uqOh0MsfvIdIQj1e6KbQRYezjWPp6Zr22EqBMXw_23w/edit?tab=t.0#heading=h.mknxt3h6z7pq
Microsoft - Full Course: What are AI agents? https://www.youtube.com/watch?v=3zgm60bXmQk&list=PLlrxD0HtieHgKcRjd5-8DT9TbwdlDO-OC&index=3
Microsoft - Full Course (Lessons 1-10) AI Agents for Beginners, https://www.youtube.com/watch?v=OhI005_aJkA
AI Agent Top Lectures, collected by Om Nalinde at https://www.linkedin.com/feed/update/urn:li:activity:7302280002853597184?commentUrn=urn%3Ali%3Acomment%3A%28activity%3A7302280002853597184%2C7302280049267793920%29&dashCommentUrn=urn%3Ali%3Afsd_comment%3A%287302280049267793920%2Curn%3Ali%3Aactivity%3A7302280002853597184%29
https://www.youtube.com/watch?v=LPZh9BOjkQs
https://www.youtube.com/watch?v=KTonvXhsxpc
https://www.deeplearning.ai/short-courses/how-transformer-llms-work/
https://www.youtube.com/watch?v=M6vZ6b75p9k&ab_channel=NarimanCodes
https://www.youtube.com/playlist?list=PLoROMvodv4rOSOPzutgyCTapiGlY2Nd8u
https://www.youtube.com/watch?v=aHCDrAbH_go&ab_channel=LangChain
https://www.deeplearning.ai/short-courses/build-long-context-ai-apps-with-jamba/
https://huggingface.us17.list-manage.com/subscribe?u=7f57e683fa28b51bfc493d048&id=9ed45a3ef6
https://www.deeplearning.ai/short-courses/collaborative-writing-and-coding-with-openai-canvas/
https://huyenchip.com//2025/01/07/agents.html
https://www.youtube.com/watch?v=KrRD7r7y7NY&ab_channel=SnowflakeInc
https://www.youtube.com/watch?v=CZ1v8uvxbL8&ab_channel=JulianGoldieSEO
https://www.youtube.com/watch?v=tx5OapbK-8A&ab_channel=DaveEbbelaar
https://www.youtube.com/watch?v=bZzyPscbtI8&ab_channel=DaveEbbelaar
0. 10 GitHub repos that can teach you everything about AI Agents by Om Nalinde, https://www.linkedin.com/posts/that-aum_10-github-repos-that-can-teach-you-everything-activity-7349397581401055234-nTJE/?utm_source=share&utm_medium=member_android&rcm=ACoAAAPsW64BNtxodzQe3M_H7WwpWaac2Y0ycAQ
1. Hands-On Large Language Models
Complete set of code notebooks from the book, covering everything from basics to advanced LLM fine-tuning.
https://github.com/HandsOnLLM/Hands-On-Large-Language-Models
2. AI Agents for Beginners
A free 11-part intro course to help newcomers build their first AI agents.
Repo Link: https://github.com/microsoft/ai-agents-for-beginners
3. GenAI Agents
Tutorials and code for building generative AI agents, ranging from simple to advanced techniques.
Repo Link: https://lnkd.in/gaiZgzpA
4. Made with ML
Learn to design, build, deploy, and improve production-ready machine learning applications.
Repo Link: https://lnkd.in/gn2RhT_m
5. Prompt Engineering Guide
Collection of resources, tutorials, and papers to master writing effective AI prompts.
Repo Link: https://lnkd.in/gDt3Sknr
6. Hands-On AI Engineering
Showcase of practical LLM-powered apps and agent-based solutions.
Repo Link: https://lnkd.in/gDM7VWcu
7. Awesome Generative AI Guide
Curated hub for generative AI research, learning materials, and tools.
Repo Link: https://lnkd.in/gsTk5nXx
8. Designing Machine Learning Systems
Summaries and resources based on the Designing Machine Learning Systems book.
Repo Link: https://lnkd.in/g26KNXfb
9. Machine Learning for Beginners from Microsoft
A free, beginner-friendly introduction to machine learning concepts and practices, provided by Microsoft.
Repo Link: https://lnkd.in/gGithBVP
10. LLM Course
Hands-on course with roadmaps and notebooks to design, build, and deploy LLM apps. Repo Link: https://github.com/mlabonne/llm-course
11. GenAI - Agent OS. This repository provides the complete infrastructure for running GenAI agents.
https://github.com/genai-works-org/genai-agentos
12. Agent-Wiz. Agent Wiz is a Python CLI for extracting agentic workflows from popular AI frameworks and performing automated threat assessments using established threat modeling methodologies.
Papers on AGI of Note | Related
Sparks of AGI (2023)
Evaluation of OpenAI o1: Opportunities and Challenges of AGI (2024)
Managing extreme AI risks amid rapid progress (2024)
An Approach to Technical AGI Safety and Security (2025)