Project Track 9: Local Heritage & Architecture Guide

The Problem / Concept A cultural exploration app tailored to Kathmandu’s specific historical sites, offering deeper context than a standard map.

Project Overview & Objectives

Standard map apps provide basic business information, but lack deep cultural and historical context for heritage sites. This app acts as a digital historian, providing nuanced explanations of local architecture, mythology, and cultural practices (e.g., Kathmandu Valley).

If you select this track, your goal is to build a functional Minimum Viable Prototype (MVP) that seamlessly integrates a Node.js/Express backend with Google’s Gemini API, utilizing Retrieval-Augmented Generation (RAG), multimodal vision, and autonomous tool calling.

Detailed Requirements Document (PRD)

1. RAG (Obsidian) Core Requirement

To prevent hallucination, the AI must be grounded in a specific, personal knowledge base. You will build this using Markdown files in Obsidian.

  • Knowledge Base Content: The knowledge base includes detailed notes on local history, the cultural significance of specific deities, festival schedules, and architectural terminology (e.g., Newari wood carving styles, pagoda tier structures).
  • Implementation Expectation: Your Express server must parse these Markdown files, generate vector embeddings using gemini-embedding-2, and perform a cosine similarity search against the user’s query. The retrieved text must be injected into the system prompt.

2. Multimodal (Vision) Stretch Goal

AI is not just text. Modern products must perceive the world.

  • Vision Use Case: The user takes a picture of a specific temple strut, a carved window, or a stone statue. The AI uses its vision model to identify the specific deity or architectural style, and retrieves the relevant mythology from the RAG database.
  • Implementation Expectation: Your frontend HTML must include a file upload input. The image must be converted to base64, sent to the /query endpoint, and passed to gemini-2.5-flash alongside the text prompt and RAG context.

3. Tool Calling Stretch Goal

Agents need to take actions in the real world or fetch real-time data that isn’t in their RAG database.

  • Tool Definition: Integrate a get_walking_distance(origin, destination) tool using a mock Maps API, or a tool that fetches the current opening hours and ticket prices for the identified heritage site.
  • Implementation Expectation: You must define a strict JSON schema for this tool and register it in your Gemini API call. When the model decides to invoke the tool, your Node.js server must intercept the request, execute a mock function, and return the result to the model for the final response.

Step-by-Step Implementation Guide

If you are using this document to prompt an AI coding assistant (like GitHub Copilot or Cursor), use the following phasing:

  1. Phase 1: UI & Mock Backend: Ask the AI to generate a vanilla HTML/JS interface with a text input, file upload, and a submit button. Connect it to an Express POST /query route that returns mock JSON.
  2. Phase 2: Vanilla Gemini Integration: Connect the Express route to the actual @google/genai SDK. Send a simple text prompt and display the result.
  3. Phase 3: The RAG Pipeline: Ask the AI to write a script to read your Markdown files, split them into chunks, and get embeddings. Update your /query route to calculate cosine similarity and inject the top matches.
  4. Phase 4: Multimodal & Tools: Finally, add the base64 image parsing to the API call, and define your function declaration for the tool-calling stretch goal.

Note: In the future, a specific branch in the course repository will be provided containing starter scaffolding tailored to this exact project track.