Hacker News Autocomplete System

Overview

This project is an autocomplete system using a trie data structure, sourced from Hacker News titles. It aims to teach the intricacies of building an autocomplete system and explores advanced concepts like Retrieval-Augmented Generation (RAG) and integration with Large Language Models (LLM).

Features

• Autocomplete Functionality: Utilizes a trie data structure for efficient autocomplete suggestions based on Hacker News titles.
• RAG System: Leverages the concept of Retrieval-Augmented Generation for enhancing information retrieval and response generation (In progress).
• LLM Integration: Uses a Large Language Model to analyze data, identify trends, and generate relevant questions. [Looking options for utilizing open-source LLMs with faster inference methods (tvm, llama.cpp, wasm)].

TODO

• Extend trie that I could store titles
• Generate embeddings
• Store those embeddings ( interested in sqlite-vss, pgvector)
• Build connections between LLMs and data source