1.Intro and Setup
What is LangChain? What kind of applications can I build with it? Let's get you started and install all the requirements.
2.Build your first Chain
Build your first chain by concatenating a prompt to a model. Also, learn how to add memory to your chains.
3.LLMS with your own data!
Integrate your own data easily with the capabilities of models such as GPT
4.Process your data
Before using your data with models, you need to process it. In this video you'll learn about text splitters, embedding models and vectorstores.
5.Build a RAG
RAG is a super powerful technique to turn a generic Large Language Model in an expert at anything you can feed it. Let's see the different kinds of RAG and let's build one too.
6.Make your RAG more reliable through Semantic Routers
Semantic Routers allow us to guide a RAG model in its answers, it's a good way to prevent allucinations or the answering of dangerous questions.
7.Different Prompting Techniques
There are many different ways to prompt your model. Let's explore zero shot, few shot and explore the tools at our disposal to execute these.
8.Vision Models
Many AI companies also have developed vision models. Let's check them out!
9.Project: Building a customer support RAG
Let's put all the pieces together and build a RAG system to answer our customers' questions.
10.Project: Deploying the RAG!
The superpower AI has over humans is its possibility to work 24/7. Let's deploy a model to run all day every day, to reap the benefits of this.
11.The next step: AI Agents
AI can do much more than just speak: it can actually perform tasks, get things done. Let's explore these capabilities through agents.