Skip to content

Aharshi3614/Devwhisper

Repository files navigation

🎙️ DevWhisper — Voice-Native Developer Experience Agent

DevWhisper is a voice-first AI agent built for developers. Instead of stopping to search through files or documentation, you just ask out loud — and it answers based on your actual codebase.


🚨 The Problem

Developers lose focus constantly. Switching between your editor, a browser, Stack Overflow, and documentation breaks the flow of thinking. Most AI tools still require you to type, copy-paste code, and wait.

DevWhisper lets you stay in flow. Ask a question with your voice, get an answer in seconds, and keep coding.


✨ What It Does

🎤 You ask a question about your code

🔍 It searches your actual codebase semantically

🔊 It responds in plain spoken English, like a senior dev sitting next to you

Example questions that work:

  • "What does the preprocess function do?"

  • "Where is the model saved after training?"

  • "How do I debug a KeyError in the pipeline?"


🏗️ Architecture

Developer speaks

Vapi — Speech to Text

FastAPI Webhook Server

Qdrant Vector Search

Groq LLaMA 3.3 70B

FastAPI sends answer back

Vapi — Text to Speech

Developer hears the response

---

🛠️ Tech Stack

  • 🎙️ Vapi — handles voice input and output

  • 🗄️ Qdrant — stores and searches code as vectors

  • 🤖 Groq with LLaMA 3.3 70B — generates the response

  • ⚡ FastAPI — receives webhooks from Vapi and orchestrates everything


🚀 How To Run It

  1. Clone this repo

  2. Install dependencies: pip install -r requirements.txt

  3. Create a .env file in the root folder:

QDRANT_URL=your_qdrant_cluster_url

QDRANT_API_KEY=your_qdrant_api_key

GROQ_API_KEY=your_groq_api_key

  1. Add your Python files to the sample\_codebase folder

  2. Index your codebase: python indexer.py

  3. Start the server: uvicorn main:app --reload --port 8000

  4. Expose publicly: ngrok http 8000

  5. Update Vapi tool Server URL with your ngrok URL plus /webhook


📁 Project Structure

  • main.py — FastAPI webhook server, handles all Vapi events

  • indexer.py — chunks your code files and uploads them to Qdrant

  • retriever.py — takes a query and finds the most relevant code chunks

  • llm.py — sends the query and context to Groq and returns the answer

  • sample\_codebase/ — put your own Python project files here


⚠️ Notes

  • The ngrok URL changes every time you restart it. Remember to update the Server URL in your Vapi tool settings each time.

  • The .env file is not included in this repo for security. You need to create your own with the keys above.


👩‍💻 Built By

Aharshi Sinha — CSE AIML

About

DevWhisper is a voice-first AI agent built for developers. Instead of stopping to search through files or documentation, you just ask out loud — and it answers based on your actual codebase.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages