Hey there, friend! It’s Daniel here, the author of 📘 LangChain for JavaScript Developers and the guy who runs the js-craft.io website.
This is the 56th edition of the Js-Craft newsletter. If you like this newsletter you can subscribe to it here.
I’m super excited to share some cool things with you, including my interview with Jacob Lee, the lead maintainer of LangChain.js. Let’s dive in!
In this edition of the newsletter:
- 🎙 Interview with Jacob Lee, lead maintainer of LangChain.js
- 🏆 Best articles from around the web
- 🎓 AI Engineering challenge – ChatGPT Vision API with LangChain and JavaScript
- 🖥 Home office setup updates
- ❓ Did you know that…
🎙 Interview with Jacob Lee, main maintainer of LangChain.js
Below is my interview with Jacob Lee, the lead maintainer of LangChain.js and a founding engineer at LangChain.
In this talk, we cover topics such as:
- How to get started with LLMs and AI models as a JavaScript developer
- What LangChain is and the benefits of using it
- A good learning roadmap for LangChain.js
And much more.
I’ve been working for two weeks on organizing and editing this interview with Jacob. I hope you enjoy it 😊
🏆 Best Articles from Around the Web
Cool things I found during my daily internet stroll:
- UI Interfaces in the Time of AI – AI is here to stay, but does that mean we can replace UIs with chatbots? Malte explains why "pure chat" isn’t enough and why we still need engaging UIs.
- Chrome Canary Now Has a Built-in LLM – Named Gemini Nano, it can be used in JavaScript via the
window.aiobject. - New JavaScript Features – The next version of JavaScript is agreed upon, and Axel Rauschmayer tells us what we need to know about the new specs.
- What Makes Great Animations? – Creating animations with CSS is easy, but Emil Kowalski says great ones feel natural, fast, interruptible, and accessible. His post is full of tips and eye candy to improve your animation skills!
- Job Hunting in Tech – Josh Collinsworth shares solid advice on applying, interviewing, and landing your next tech job.
🎓 AI Engineering Challenge – ChatGPT Vision API with LangChain and JavaScript
As you may have seen in previous editions of this newsletter, I’m showcasing scenarios for using AI models in web applications with JavaScript.
In the previous editions, we have seen how to use LangChain to execute and generate JavaScript code and how to make a full synonyms app with ChatGpt and ReactJs.
Today’s example focuses on using the ChatGPT Vision API with LangChain.js. This has been one of the most enjoyable examples I’ve worked on recently. It’s very easy to use and holds great product potential.
Compare this with using TensorFlow.js to identify objects in a picture. It’s much simpler 😊
🖥 Home Office Setup Updates… kind of
Last week, I spent time with my wife and daughter at the seaside on a "workation." I was lucky to have this view from my desk.
I learned that:
- I can work much better with good relaxing activities.
- A nice view is really motivational.
- The NexStand laptop stand is the best investment I’ve made for my cervical back pain and posture.
More and more, I’m starting to believe that big, crowded cities are overrated and can actually be an enemy of happiness.
❓ Did You Know That…
… the protocol HTTP/0.9, the first version used on the World Wide Web, was incredibly simple compared to today’s standards? It only allowed for one command: GET, which requested a webpage from a server. There were no headers, cookies, or any other type of metadata—just the raw HTML document. This simplicity was key to the Web’s initial growth, allowing for easy implementation and widespread adoption.
What did you think of today’s newsletter?
- ⭐️⭐️⭐️⭐️⭐️ Loved it!
- ⭐️⭐️⭐️ Room for improvement.
- ⭐️ Not my jam.
That’s all for this edition, folks! Be kind, be smart, and keep coding!
Wish you all the best,
Daniel
📖 Neural Networks from Scratch - Presale
I'm writing a book about the timeless foundational concepts of neural networks for JavaScript developers. Go from if-else to weights and biases by building tiny AI models from scratch!
📖 Neural Networks from Scratch - Presale
I'm writing a book about the timeless foundational concepts of neural networks for JavaScript developers. Go from if-else to weights and biases by building tiny AI models from scratch!