AI Courses Review: From ChatGPT to LangChain, Learning the Skills That Shape Tomorrow
Dive Deep into AI Development: Unlock the Future of AI: Master Cutting-Edge Technologies with Free Courses
Ready to dive into the world of AI development? I have great news for you!
DeepLearning.AI offers more than 20 completely free short courses designed to introduce learners to the hottest topics in AI today, including Large Language Models (LLMs), ChatGPT, LangChain, Text Embeddings, Vector Databases, ML Ops, and more. These courses provide an excellent gateway into the AI world for individuals ranging from beginners to more experienced software developers.
Unlock the full potential of AI with “Practical Prompt Engineering” — featuring 250+ expert-level prompts and comprehensive guides for ChatGPT and other AI tools. Includes industry-specific templates for finance, healthcare, and manufacturing. Sample the first chapter at futureproofskillshub.com/prompt-engineering. Available in paperback on Amazon, and as an ebook on Apple Books, Barnes & Noble, and Kobo.
Master the complete tech stack at futureproofskillshub.com/books — from AI to Python, SQL, and Linux fundamentals. Plus, discover how to maintain peak performance and work-life balance while advancing your technical career in “Discover The Unstoppable You”.
All courses include code in interactive Jupyter notebooks and videos. You will get the opportunity to explore various use cases and gain hands-on experience with these technologies. The code examples are provided in Python and JavaScript.
I highly recommend these courses for numerous reasons: the quality of the material is exceptionally high, the instructors are top-notch, and they come from companies that have pioneered these technologies.
I have completed 6 of these courses and looking forward to share my review and experiences. My hope is that this will encourage you to explore these courses and acquire some of the most sought-after skills in the market today. Learning new skills is always fun and it also helps us to stay competitive in the job market.
I am sharing all the code in Jupyter notebooks and my detailed notes on each chapter of every course in my GitHub repository.
Below is the list of courses I will be reviewing. This newsletter includes reviews for only the first 3 courses. The reviews for the remaining 3 courses can be found in the "Part 2" of the reivew:
ChatGPT Prompt Engineering for Developers
Building Systems with the ChatGPT API
LangChain: Chat with Your Data
LangChain for LLM Application Development
Building Generative AI Applications with Gradio
Build LLM Apps with LangChain.js
1. ChatGPT Prompt Engineering for Developers
If you are new to LLMs and ChatGPT, I recommend taking this course first.
The "ChatGPT Prompt Engineering for Developers" course is all about teaching developers how to use Big Language Models (LLMs) to make software. It focuses on how to give these LLMs instructions in a clear way so they can be used to do things like summarise text, figure out information, change text, or even build chat systems. These special LLMs are good for making practical apps because they are designed to be helpful, honest, and safe.
Here are the main topics the course covers:
Best Practices for Using LLMs: It talks about how to properly ask LLMs to do tasks, using examples like making summaries, changing text styles, and more. The course stresses the importance of giving very clear and detailed instructions to get the best results.
Understanding LLMs: There's a part that explains the difference between regular LLMs and those that are specially tuned to follow instructions better. This helps in making apps that work well and give reliable responses.
Developing Effective Prompts: Like in machine learning, developing prompts is a process where you start with a basic idea and keep improving it based on how well it works. This part is all about refining your questions to get better answers from the LLM.
Practical Uses: The course shows how LLMs can be used for a bunch of different things, from quickly understanding big chunks of text to creating detailed responses based on short instructions. It highlights using LLMs responsibly and tailoring the output for specific needs.
Building Chatbots: A big part of the course is about how to create chatbots using LLMs. It covers everything from setting it up to managing the flow of conversation. It even includes tips on how to make the chatbot's responses more predictable or creative, depending on what you need.
In summary, this course is great for developers looking to learn how to make their software smarter with the help of language models. It covers everything from the basics to more advanced topics like custom chatbots, making it a comprehensive guide for starting using LLMs in software development.
2. Building Systems with the ChatGPT API
This is the right course to take after the ChatGPT prompt engineering course.
The "Building Systems with the ChatGPT API" course teaches developers how to create advanced apps using Big Language Models (LLMs), with a special focus on the ChatGPT API. It starts with simple instructions and moves to building complicated systems that need to work step-by-step with LLMs. A key example in the course is creating a customer service assistant that shows how to manage user questions, analyse content, find and get the right info, and provide correct answers.
Here's a breakdown of the main parts of the course:
LLM Basics: It covers the basics like how LLMs learn, how to make them follow instructions better, breaking down text for processing, and the format for chat conversations. These are important to know how LLMs work and how to make apps with them.
Security Practices: The course stresses the importance of safely managing API keys, which are needed to use the ChatGPT API without risks.
Handling Complex Tasks: It introduces "prompt chaining," a way to tackle big tasks by breaking them into smaller, easier steps. This helps make the whole process more manageable and accurate.
Evaluating Inputs: A big part of the course is about checking inputs — classifying them, moderating for safety, and using strategies to keep the system from being tricked or misused. It talks about using OpenAI's tools for keeping content appropriate and processing inputs smartly.
Checking Outputs: It also covers how to make sure the answers or responses the system gives are good, relevant, and safe. This includes using tools to check content and asking the model itself to evaluate its responses.
Building a Complete System: The course walks you through making a full system, from receiving a user question to pulling up product info and answering. This part is hands-on, showing how everything learned can be used in a real project.
Improving Outputs: Finally, it discusses ways to make the app's responses better over time, using a step-by-step approach to refine answers, similar to how you'd improve a machine learning model.
The course is packed with useful information, but it's best for those who already know a bit about AI and coding. It's highly recommended for anyone looking to learn how to build solutions with ChatGPT API.
3. LangChain: Chat With Your Data
If you want to learn how to enhance your AI applications beyond pre-trained model capabilities, then this is the right course for you.
The course "LangChain, Chat with Your Data" teaches you how to create chatbots that can talk about personal or private data. Large language models like ChatGPT can't do it on their own because they're limited to what they learned up until their last training update. Using the open-source framework LangChain, the course teaches how to build chatbots that interact with current or confidential documents, covering everything from loading documents to making a chatbot remember past interactions.
Key teachings of the course include:
Document Loading: It begins with how to load various types of data (like web pages, PDFs, and YouTube videos) into a format that chatbots can use. LangChain supports over 80 types of document loaders, making it versatile for different data sources.
Document Splitting: This part explains the importance of breaking down large documents into smaller pieces for better information retrieval. It covers how to split texts while keeping their meaning intact and retaining useful metadata.
Vectorstores and Embeddings: After splitting, documents are indexed using embeddings in a vector store, making it easier to find relevant pieces of text for answering questions. This step is crucial for the semantic search that powers the chatbot's understanding.
Retrieval: The course goes into advanced retrieval methods to find the most relevant and diverse information from the stored data. Techniques like Maximum Marginal Relevance (MMR) and self-query retrieval are introduced to enhance search results.
Question Answering: With the right documents retrieved, the course shows how to use them to answer user questions. It addresses challenges like managing large volumes of data and refining answers for better accuracy.
Chat: The final lessons focus on building a chatbot that can handle follow-up questions, using chat history to provide context-aware responses. This involves setting up a conversational retrieval chain and integrating various components for a seamless chat experience.
In short, this course emphasises practical applications, from handling YouTube transcriptions to dealing with proprietary databases. It's a hands-on journey from loading and processing documents to creating a chatbot capable of meaningful, data-driven interactions.
Wrap up
These courses dive into some of the coolest AI tech out there, from the basics of Large Language Models (LLMs) and ChatGPT to the more advanced LangChain framework. Whether you're new to AI or you've been coding for a while, there's something in these courses for you.
The course on LangChain really stood out to me. It shows you how to get around the usual limits of LLMs, like ChatGPT, especially when you need to work with the latest data or private documents. This is super useful for building chatbots that can chat about specific, up-to-date info.
If you're keen on getting into AI or levelling up your skills, definitely check out these courses. They've been a game-changer for me, and I believe they can be for you too. Trust me, diving into these courses is a step toward mastering some of the most valuable skills in the market today.
Don’t forget to check out my next newsletter where I will review these 3 courses:
LangChain for LLM Application Development
Building Generative AI Applications with Gradio
Build LLM Apps with LangChain.js
Happy learning!
Btw, feel free to leave a comment or to describe your learning path in the AI domain!