Hello everyone and welcome to this video

on agentic AI frameworks. It seems like

everywhere you look today, businesses

are turning to AI agents to automate

complex workflows, boost productivity

and make smarter decisions. Across

industries like finance, healthcare

retail and tech, more and more companies

are already using AI agents to handle

tasks that once required entire teams of

people. Maybe you're here because you

have heard the buzz around agentic AI

and want to finally understand what it

actually means. Or maybe you're curious

about which framework is best to build

your own intelligent systems. And you

might be wondering what exactly is an

agentic AI framework? How is it

different from traditional AI tools? And

how can you use it to build autonomous

AI systems that don't just respond but

actually think, plan, and act? In this

video, we are going to break it all down

step by step. From understanding what AI

agents really are to exploring popular

frameworks, their key features, how to

choose the right one, and where they are

used in real world. Whether you are a

developer looking to build powerful

agents or a business leader exploring

automation, by the end of this video

you will have a clear road map to

agentic AI frameworks. Now, before we

dive into the real world example, please

like, share, and subscribe to Edureka's

YouTube channel and hit the bell icon to

stay updated on the latest content from

Edureka. Also, check out Edureka's

agentic AI certification training. It is

carefully crafted to meet industry

demands and prepare you for the future

of intelligent agents. You will gain

practical skills in lang rag LLM ops and

more through live instructorled sessions

and hands-on labs. Whether you're a

beginner or a tech professional, this

course helps you master the concepts and

accelerate your AI career. Now, let's

take a quick look at how AI agents are

making real impact. Starting with a

powerful example in customer experience.

Imagine reaching out to customer support

and getting instant personalized help.

No repeating details. That's what

agentic AI brings to customer

experience. With rising customer

expectations and growing burnout among

support teams, AI agents are stepping in

to make every interaction smarter and

faster. These agents don't just respond

they learn, remember, and act. Unlike

traditional chat bots that follow

scripts, Agentic AI understand context

predicts needs, and can even take

proactive action like offering a refund

opening a support ticket, or escalating

an issue before it becomes a complaint.

They use natural language processing to

hold real conversations, sentiment

analysis to sense emotions, and can

smoothly hand over complex cases to

human agent when needed. And behind the

scenes, these agents assist customer

service representatives too fetching

data, troubleshooting problems and

suggesting solutions in real time

because they can interact with multiple

system and remember customer details.

Agentic AI delivers supports that not

just quick, it's deeply personal and

proactive. And the result is happier

customers, reduce workload for agents

and improved efficiency for businesses.

And all powered by intelligent evolving

AI systems. When most people hear AI

agent, they picture a simple chatbot.

But real AI agent go far beyond just

replying to questions. An AI agent is a

system that can understand a goal, plan

how to achieve it, act on that plan, and

learn from the results without constant

manual input. Let me break this down.

First, the agent understands the task.

For example, if you say, "Give me a

daily sales report," it knows it needs

numbers, trends, and summaries. Next, it

creates a plan like pulling data from

your CRM, cleaning it, and generating

insights. Then, it connects to external

tools such as APIs, databases, web

searches, or even other AI agents to

gather what it needs. After that, it

executes the plan step by step. And

finally, it learns from the feedback

storing that experience in memory to

perform better next time. So, that's the

real power of AI agents. They don't just

react, they reason, act and improve over

time. Now, here's the challenge.

Building such a capable agent from

scratch is not easy. You would have to

design its architecture, handle

communication between components, manage

memory, integrate external tools, and

make sure everything runs smoothly.

That's a lot of time, effort, and

maintenance. Agelic frameworks solve

this problem by giving developers and

organizations a readymade structure to

build agents like how a game engine

saves developers from writing the entire

physics of a game from scratch. So think

of it like this. An AI agent is like a

skilled driver. An agentic framework is

like the root system providing

direction, structure, rules and smooth

connections between destination. So

without a framework, your agent can

work, but it's slow, fragile, and hard

to scale. With a solid framework, it can

move fast, connect to multiple tools

and collaborate with other agents

effectively. So I hope now it's clear to

you all why we need agentic frameworks.

Now, a good agentic framework isn't just

a tool. It's a complete environment that

handles the heavy lifting behind the

scenes. And here are some of the most

important features. First is the defined

architecture, a clear blueprint for how

the agent plans, decides and interacts.

Next comes the communication layer. It

enables smooth interaction with APIs

databases, humans and other agents. Next

is the task management. It handles

multiplestep task, priorities, and

dependencies. Next comes the tool

integration. Pre-built connectors make

setup faster and easier. Next is memory

and learning. It allows agents to

remember past interaction and improve

over time. And finally is the monitoring

and control. It gives visibility into

what the agent is doing making it easier

to debug and optimize. So in short I can

say that the framework gives the

foundation so you can focus on building

intelligence instead of worrying about

infrastructure. So why are these

frameworks gaining so much attention?

because they allow businesses to scale

AI beyond single isolated task. With

agentic frameworks, organizations can

build multiple agents that collaborate

on complex workflows, automate entire

processes end to end, keep systems

stable and consistent as they grow, and

deploy solutions faster since the core

infrastructure is already in place. So

this is what turns a simple chatbot

experiment into a real AIdriven

operation. Now let's look at some of the

most popular agentic frameworks today.

So first on the list we have lang chain.

It's ideal for connecting language

models with external tools and creating

multi-step reasoning pipelines. Lang

chain is an open-source framework that

helps developers build application using

large language models like GPT clar or

llama. These apps can think use tools

and remember context. It connects LLN to

things like APIs, databases, documents

computational tools, and memory system

letting them follow multi-step logic.

Basically, Lanchin turns an LLM into an

intelligent agent that can perceive

plan, and act. With LChain, you can

build agents fast and your way. Use

readymade templates and patterns like

React to create agents in minutes. swap

models, tools or databases easily with

over a thousand integrations. You can

also customize agents with a simple

middleware for approvals, conversational

management or handling sensitive data.

And with Lang Graphs durable runtime

your agents get persistent checkpoints

and human in loop support automatically.

So with that said, next we have

Langraph. It is powerful for designing

complex workflows where multiple agents

interact. Keep your agents on track with

human in the loop checks and easy

moderation. You can guide, approve, and

control what your agents do. Langraph

makes it simple to build customizable

workflows, whether it's a single agent

multiple agents, or hierarchical setup.

It also remembers conversations for

richer long-term interaction. Plus, with

real-time streaming, users can see what

the agent is thinking and doing, making

the experience smooth and interactive.

The next framework on our list is

Autogen. Great for building multi- aent

systems that collaborate like a team.

So, Autogen is an open-source framework

by Microsoft that lets you create AI

agents that can collaborate not just

with humans, but also with other AIs.

These agents can chat, reason, plan

write code, use tools, and even review

each other's work to finish complex task

automatically. Autogen works like a

team. The user proxy agent represents

the user. It gives instructions or goals

and the assistant agents plan, reason

and execute task. And a critic agent can

review the output and suggest

improvements. So these agents

communicate through messages just like

people chatting in a group project

discussing steps until they reach the

right answer. Autogen makes it easy to

build multi- aent systems for research

automation, software development, data

analysis, content generation. And it

saves time, reduces human effort and

improve quality through self-re and

collaboration. So you can create an

autogen setup where one agent writes

code, another test it and third reviews

it all automatically. Autogen turns AI

models into autonomous collaborators

that can think, talk and work together.

It's like giving your AI teammates each

with their own role to solve problems

faster and smarter. Next on our list, we

have crew AI. It focuses on

orchestrating specialized agents to work

together on complex task. Crew AI makes

it easy to build and manage

collaborative AI agents that can handle

complex task on their own and at scale.

It's easy, trusted, and scalable

helping businesses adapt AI across teams

with centralized management and

monitoring. You get LLM and tool

configuration, role- based access and

serverless containers. Next, moving on

to open dating built for developer

agents. It enables coding AIs that can

write, test, and deploy code

independently. Open Daven is an

open-source project aimed at creating an

autonomous software engineer, an AI

agent that can understand software task

write code, debug, test, and deploy

solutions automatically. Open Daving

uses large language models combined with

specialized tools and environments to

perform end-to-end software development

task. It can understand developer

instructions, plan coding step, write

and modify code, run and test the

scripts, fix bugs and deploy solutions.

Basically, it acts as an AI pair

programmer or even a full autonomous

coding agent. Open Dev is part of the

new wave of agentic AI frameworks

systems that can act, learn and

collaborate. It helps developers

automate repetitive coding task, debug

faster, build prototypes independently

and finally improve productivity. Now

let us move on to next framework. We

have semantic kernel. A lightweight and

flexible framework for integrating

external skills easily. Semantic kernel

is an open-source toolkit that makes it

easy to build AI agents and connect the

latest AI models to your C, Python, or

Java projects. It works like smart

middleware, turning model requests into

function calls and sending results back

fast. You can plug in your existing code

as extensions, integrate AI services

easily, and share them across your team.

It's modular, flexible, and built to

rapid enterprise solutions. The next

framework on our list is Llama Index. It

is perfect when your agent need to work

with structured data. Llama index is an

open-source data framework that helps

developers connect large language models

like GPT or Llama to external data

sources such as databases, PDFs

documents, APIs or websites. LLMs are

great at reasoning and generating

language, but they don't naturally have

access to your private data like

internal documents, business reports, or

real-time information. Lama index acts

as a bridge between LLMs and your data.

It ingests data from any source like

text files, PDFs, SQL databases, APIs

notion, etc. Indexes that data

efficiently for retrieval. feeds the

most relevant information back to the

LLM when you ask a question. So this

process is known as rack which is

retrieval augmented generation. So each

framework brings something unique. Some

emphasizes tool integration other

collaboration or data handling. The

right choice depends on your use case

and goals. So there is no single best

framework only the best one for your

specific needs. So, here's a simple way

to choose the right agentic framework.

First, start with your goal. Are you

building a chatbot, an autonomous flow

or a multi- aent ecosystem? Next, check

integration needs. Can it connect easily

to your tools and data sources? Then

consider scalability. Think, will you

need more agents later? Next, look for

flexibility. Can you customize how your

agents reason and act? and then evaluate

community support. Good documentation

and an active community save time. Then

balance cost and performance. Choose

something that fits your resources and

growth plans. So your framework should

align with the problem you want to solve

and not the other way around. So there

is a common confusion for beginners. AI

agent builders are like a readytouse

kit. You can drag, drop, and launch an

agent fast, which is perfect for simple

task like a support bot, but they are

limited in flexibility. Agentic

frameworks, on the other hand, give

developers the freedom to design

powerful customized systems from the

ground up. So, it's just that the

builders are like instant cake mix, fast

but limited. Frameworks are like having

all the ingredients. It takes more

effort, but you control the flavor

shape, and result.

So, agentic frameworks are becoming the

backbone of modern AI automation. They

make it possible to build intelligent

autonomous agents that don't just

respond but plan, act, and learn. As

agentic AI continues to grow, this

frameworks will play a key role in

shaping the future of automation. If

this video helped you understand the

concepts clearly, don't forget to like

share, and subscribe for more deep dives

on AI, agentic systems, and genai tools.

And with this we have come to an end to

this video on agentic AI frameworks. If

you enjoyed listening to this video

please be kind enough to like it and you

can comment on any of your doubts and

queries. We will reply to them at the

earliest. And do look out for more

videos and playlist and subscribe to

Edureka's YouTube channel to learn more.

Thank you for watching and happy

learning.