ReAct Prompting with LangChain

This post is somewhat outdated considering how quickly generative AI technology is advancing. I have to confess I wrestled with procrastination for a few months before finally getting it published 😊. But, anyway, here we go!

For some time, ChatGPT screenshots were all over my social media feeds, with some good-faith users demonstrating its impressive capacity to surpass human intelligence, while others made attempts to poison the prompt and reveal ChatGPT’s limitations. Shortly after, GPT4, GPT plugins, and a whole bunch of new development tools sprung up. Although the dynamics were exciting, it felt like there was always something new to learn, which is why I used our innovation sprint to get a better grasp of the hype (Surprise! I was programming in Kotlin since I joined Atlassian! Who would have thought an ML engineer would be dabbling in a non-Python language!)

In the upcoming sections, I share back some of the high-level insights.

1. A brief overview of ChatGPT

1.1 ChatGPT Mega hit

In late November 2022, OpenAI released ChatGPT as a research preview that comes with a user-friendly conversational interface, and within only a few days, ChatGPT became a huge hit.

ChatGPT is not open-sourced, so we can’t be entirely sure how it was developed. However, we do know that ChatGPT is fine-tuned from a model in the GPT-3.5 series, which, in turn, is an upgraded version of GPT-3 that was released in 2020.