I recently watched Andrej Karpathy’s video lecture “Let’s build GPT: from scratch, in code, spelled out.” found here: Youtube

Previously, we discussed building server side to support creating a ChatGPT like Q&A for a private or proprietary text document. In this final blog for this series, we will discuss client side solutions to complete this.

In the previous blog, we discussed building embeddings for custom documents and saving embeddings to PineCone vector Database. In this blog, we will continue on to build an end-to-end chat solution using LangChain library, PineCone service with the code to be hosted on AWS Lambda function and APIs exposed through API Gateway.

ChatGPT is an AI chatbot built with Large Language Model (LLM) that can converse in a human like manner. ChatGPT can do a lot of natural language processing activities like answer questions, generate content, translate text to a different language etc.

Hello, world! Welcome to my digital nook where I pour out my thoughts, ideas, and experienences related to software development. My name is Sekhar Kafle, a seasoned software engineer, an IT leader, and a passionate enthusiast of Artificial Intelligence and Machine Learning. In this multidimensional playground of mine, you’ll get a firsthand perspective into my explorations across these diverse but intertwined domains.