Top 5 Youtubers That Will Teach You LangChain

When it comes to learning a new topic, everyone’s wired differently; some are perfectly content to squat in theory-land indefinitely, some enjoy grappling with theory for a few rounds before hobbling together a pragmatic application, and some prefer learning Bob Ross style—which is to say watching someone do the thing you’re interested in learning and following along.
​
For this last option, YouTube is a wonderful resource. Particularly for newcomers to the field, when a new machine learning framework arrives, it can be motivating to see someone else paint out for you what that framework can do before you slog through its documentation.
​
Released in October 2022 and rapidly evolving since, the LangChain framework is designed for hacking together applications that connect large language models (LLMs) with other tools at fairly high levels of abstraction, ideally giving rise to more complex functionality than an LLM can offer alone. Though they make it powerful, LangChain’s numerous components—schemas, models, prompts, indexes, memory, chains, and agents—can make LangChain a lot to wrap your head around.
​
Thankfully, via videos filled with examples, explanations, and walkthroughs that’d make Bob Ross proud, several YouTubers are helping folks get their feet wet with LangChain. In no particular order, let’s go over the top five YouTubers for learning some LangChain.

​First up is a channel named after its creator, the Singapore-based CEO and co-founder of Red Dragon AI, Sam Witteveen. Sam’s LangChain videos cater to a range of experience levels but, ideally, you’ll want some coding background (he uses Python) and familiarity with Jupyter Notebooks to follow along. Even if you’ve never coded or used Jupyter Notebooks, though, you can still gain a decent sense of LangChain’s capabilities and limitations from Sam’s videos because he shares copious concrete examples of where various LangChain applications get things right and wrong.
​
Easily the most comprehensive LangChain channel on this list, Sam’s LangChain playlist is comprised of 33 videos, ranging from a few beginner LangChain videos to more complex applications based on academic papers, like this one that uses LangChain to build an app akin to Microsoft’s Visual ChatGPT. Sam covers LangChain applications that use closed LLMs (primarily OpenAI’s) and open LLMs (like this one that uses Alpaca-7B). Below one of Sam’s tutorials showing how to make a LangChain app that employs several tools:

Sam’s videos are detailed, hold-your-hand walk-throughs of different LangChain features and applications, interspersed with conceptual explanations and occasional sections of academic papers. You can open up any video’s accompanying Jupyter Notebook (linked to in his video descriptions or found in his GitHub repo) and follow along. Sam also actively responds to questions and suggestions in his comments section.

​The next channel is named after its creator, James Briggs, the Dubai-based founder of Aurelio AI and developer advocate manager at the vector database company Pinecone.
​
James’ channel currently contains 13 LangChain videos, mostly ironing out the LangChain basics and core components. He also wades into more complex LangChain topics, like the one below, where he builds conversational agents that use vector databases. James’ LangChain tutorials utilize both the closed OpenAI LLMs and actual open LLMs, like Falcon 40B (currently ranked highest on Hugging Face’s Open LLM Leaderboard).

James’ LangChain series is beginner-friendly with the same caveats that Sam Witteveen’s channel had (you’ll want some coding and Jupyter Notebook familiarity). James’ videos also link to their accompanying Jupyter Notebooks, stored at this GitHub repo, so you can follow along.
​
Likely due to his developer advocate experience, James shines in not glossing over details that are simple to some folks but that could easily leave others lost (e.g., generating a Hugging Face or OpenAI API key). Another nice touch is that James breaks up his code walkthroughs with short conceptual explanations, stock video clips, and sections of relevant academic papers, so you’re not staring at Jupyter Notebooks for entire videos.​

​Based in San Francisco, Greg Kamradt brings his background in product development, data engineering, and data analytics to the next LangChain channel on our list (named after Greg).
​
Greg currently has 24 videos in his LangChain playlist, starting with a LangChain 101 series that contains an overview, a quick start guide, “Explain it to me like I’m five” style LangChain Cookbooks (part 1 and part 2), and introductory videos that run through applications like "Agents Overview + Google Searches" and "Question A 300 Page Book (w/OpenAI + Pinecone)." He also has more advanced videos like the one below on topic modeling from video and audio with LangChain:

Greg’s style is not dissimilar from Brigg’s and Witteveen’s in that many of his videos include Jupyter Notebook walkthroughs (in Python) of various LangChain applications. Some, however, are walkthroughs of code you’ll need to pull from his GitHub repo. For these, basic familiarity with Git will be helpful. So you can easily find an entry-point video appropriate to your experience level, Greg labeled each tutorial at his LangChain repo with categories and difficulty ratings.
​
An underlying focus on potential business applications and workflow optimizations pervades Greg’s LangChain videos. In fact, he states explicitly that he’s not interested in theory, preferring real-world applications. So if you want to get your entrepreneurial gears turning, you’ll dig Greg’s channel.​

​Abdul Majed Raja, a professional data scientist based out of Bengaluru, India, runs the next channel, 1littlecoder.
​
1littlecoder jumps into LangChain without explaining it, so if you’re starting from scratch, it’s probably better to start with one of Witteveen’s, Briggs’, or Kamradt’s introductory videos (or skim through LangChain’s intro documentation).
​
1littlecoder currently has an 8-video LangChain playlist covering some interesting LangChain applications like using LangChain to chat with Excel or CSV files, and the below video shows how to make a “NavalBot” (or bot from another personality) with LangChain and EmbedChain:

Abdul’s videos follow two main styles: Some videos are purely conceptual reviews, often of newer applications, like this one that explained BabyAGI shortly after its release. Other videos start by demonstrating what some LangChain application does, reviewing the app’s broad concepts, and then walking you through an accompanying Jupyter Notebook (again, in Python) linked to in his video descriptions (like this one that combines BabyAGI with LangChain agents).​

​Our final channel, Chat with Data, is run by the London, UK-based Founder and CEO of Sienna Analytics, Mayo Oshin.
​
Chat with Data currently contains seven JavaScript and TypeScript-based LangChain videos, including one with Harrison Chase, LangChain’s Co-Founder and CEO. Again, though, as long as you’re familiar with some coding language, you can still learn from Mayo’s front-end-focused LangChain channel.
​
Mayo initiates his series with a comprehensive beginner JavaScript/Typescript-based LangChain tutorial. It’s about an hour long (many of his videos are meatier than the above LangChain YouTubers) but has plenty of timestamps in case you prefer to chunk it up. After this tutorial, Mayo explores several more detailed LangChain applications, like using GPT-4 and LangChain to chat with a single PDF or with multiple PDFs, or the below video that walks through making a chatbot for a website using LangChain, Typescript, and Supabase, a Postgres database that can store embeddings.

Chat with Data's videos normally start off with a diagram, illustrating an app’s main components, functions, and information flows. Then, rather than walk you through a Jupyter Notebook, Chat with Data walks you through code that you’ll need to pull from one of his accompanying GitHub repos (linked to in his video descriptions). After a brief overview, expect a demonstration of the app and a deeper dive into the details. Since Mayo sometimes assumes his viewers know some key parts of the applications he builds (e.g., setting up a Supabase database), his videos are more geared toward viewers with at least intermediate experience.​

​Alright, we’ve now gone over enough LangChain YouTubers to get you started learning (and keep you busy for a while). I find it useful to burn through these videos at 1.5x speed for a quick overview and then take a crack at their accompanying code on my own, referring back to the video when something’s unclear (at regular speed). You may find another approach works better for you (e.g., going through a Jupyter Notebook on your own and then watching its accompanying video to fill in the gaps).
​
Whatever style you choose, you have a few options: you can pick a Python-based or a JavaScript-based channel; a channel that walks through LangChain’s components; a channel that assumes you already have a handle on those; a channel that explores potential LangChain business applications; or a channel that just explores new LangChain apps.
​
And undoubtedly, many more talented YouTubers talking about LangChain will spring up as the framework grows, gifting us even more learning options. If you have other LangChain educators you’d like to share (YouTubers or other mediums), drop a note on Deepgram’s Github Discussion page! https://github.com/orgs/deepgram/discussions