© 2026 Chiranjeev (@chiranr19) — All Rights Reserved. This project is source-available for viewing only; it is not open source. No copying, reuse, modification, deployment, or redistribution of any part of it (or its underlying ideas) without prior written permission — see LICENSE and SIGNATURE. Prospective employers and collaborators are welcome to read the code. · authorship sigil 4DBH·QKSW·HGOW·3DHK

Chainlit · GCP · Vertex AI — RAG System

This repository contains code that enables users to upload files via a Chainlit interface, store those files in Google Cloud Storage (GCS), and leverage Vertex AI Embeddings and Chroma DB for retrieval-augmented generation (RAG). This allows users to ask questions related to the contents of the uploaded files.

Architecture

Two paths: ingestion turns an uploaded PDF into searchable vectors; Q&A retrieves the relevant chunks and grounds the LLM's answer in them.

flowchart LR
    subgraph Ingest["Ingestion"]
        direction TB
        U["User uploads a PDF"] --> CL["Chainlit UI"]
        CL --> GCS[("Google Cloud<br/>Storage")]
        GCS --> RD["Load + split<br/>(gcs_filereader)"]
        RD --> EMB["Vertex AI<br/>Embeddings"]
        EMB --> CH[("Chroma DB<br/>vector store")]
    end
    subgraph QA["Retrieval-augmented Q&amp;A"]
        direction TB
        Q["User question"] --> RET["Retriever"]
        RET -->|top-k chunks| PR["Prompt template<br/>(LangChain)"]
        PR --> LLM["LLM"]
        LLM --> ANS["Grounded answer"]
    end
    CH --> RET
    ANS --> CL

Loading

Key Components

Chainlit: Provides an interactive interface for users to upload files and interact with the system. Google Cloud Storage (GCS): Used to store the uploaded files. Vertex AI Embeddings: Embeds the contents of the files into vector representations for retrieval. Chroma DB: A vector database used for storing and searching document embeddings. Retrieval-Augmented Generation (RAG): Allows users to ask questions about the uploaded files and retrieves the most relevant information. Features File Upload: Users can upload a file (e.g., a PDF) through the Chainlit interface. GCS Integration: The file is uploaded to a specified Google Cloud Storage bucket. Vector Embeddings: The contents of the file are embedded using Google Vertex AI Embeddings and stored in Chroma DB. RAG for Q&A: Users can ask questions, and the system retrieves relevant information from the uploaded file using Retrieval-Augmented Generation. Prerequisites Python 3.10+ Chainlit: An interactive chatbot interface. Google Cloud SDK: Set up for accessing GCS and Vertex AI. Chroma DB: A vector database. OpenAI API: For natural language processing tasks.