Fullstack Task

Fullstack Task (1)

Uploaded by

pradyukandala01

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

47 views2 pages

Fullstack Task

Fullstack Task (1)

Uploaded by

pradyukandala01

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

You are on page 1/ 2

AI-Powered Chat Interface for Querying PDF Documents Using Vector Similarity Search

Objective: Develop a web-based chat interface that enables users to query content extracted
from PDF documents. The system should use AI to generate contextual embeddings for text
segments and employ vector similarity search to find the most relevant responses. The
interface will be developed using Next.js 14.

> Note that many of the tools and libraries used in this task there are Python and TS libraries
- Always use TS.

Part 1: AI Model Development and Vector Similarity Search

- PDF Content Extraction:

- Utilize a robust PDF parsing library to extract text from documents, ensuring
clean and pre-processed output for accuracy in content retrieval.

- Embedding Generation:
- Employ a tool like Langchain for prompt generation, creating contextual
embeddings for text segments. Fine-tune this model on your PDF data to ensure
high relevance and contextuality in embeddings.

- Vector Database Integration:

- Leverage Pinecone (or a similar vector database) to store embeddings,
implementing vector similarity search for efficient and relevant text segment
retrieval in response to queries.

- API Development:
- Design a RESTful API that interfaces with the vector database, fetching and
serving text segments from the PDFs based on user queries with robust error
handling.

Part 2: Chat Interface Development Using Next.js 13

- Chat UI Development:
- Create a responsive chat interface using Next.js 13, enabling real-time user
queries and displaying API responses in a conversational format.

- Integration with AI API:

- Ensure seamless communication between the chat interface and the AI backend,
displaying relevant PDF text segments as chat responses.

- User Feedback Integration:

- Incorporate a mechanism for users to provide feedback on response relevancy
and accuracy, which can be used for future improvements.

Part 3: Continuous Improvement and Learning

- Model Refinement:
- Utilize user feedback and interaction data to refine the embedding model,
adopting an iterative improvement approach for enhanced accuracy over time.

- Chat UI Enhancement:
- Continuously explore Next.js 13 features to enhance the chat UI, focusing on
user engagement, accessibility, and convenience features like voice input.

Deliverables:

- A Github repo
- A Loom.com video explaining what you built with Camera and Audio Voiceover
- Chat Web Application: A fully functional chat interface built with Next.js 13, including
user documentation and challenges faced. - Video
- Project Report: Detailed documentation covering the project's methodology, technology
stack, encountered challenges, and key learnings in the Github Repo Readme