DEV Community

Kuldeep Dhanora
Kuldeep Dhanora

Posted on

A Patient Practice Partner for a Friend

This is a submission for the Hacktoberfest Weekend Challenge: Build for a Friend

What I Built

I built a friendly AI-powered language learning web app that helps people practice speaking and writing without being afraid of making mistakes.

The app works like a patient practice partner. It:

💬 Provides conversational practice
✏️ Gently corrects language mistakes
💡 Gives simple explanations and useful tips
😊 Encourages the learner instead of judging them
🎯 Supports different learning levels
🌍 Supports English, Spanish, and Hindi
📊 Keeps track of practice progress
🤖 AI / Open Innovation

The app uses the open-weight openai/gpt-oss-20b model through Hugging Face Inference Providers.

I chose an open-weight model because it makes AI experimentation more accessible and gives developers more flexibility when building their own applications.

🛠️ How I Built It

Technologies used:

Python
Streamlit
Hugging Face
openai/gpt-oss-20b
Hugging Face Inference Providers

The application is deployed as a live Streamlit web app.

Demo

https://patient-language-practice-partner-gje4qeu3l3xp5exymaatyf.streamlit.app/

Code

https://github.com/kuldeepdhanora40-prog/patient-language-practice-partner.git

How I Built It

Technologies used:

Python
Streamlit
Hugging Face
openai/gpt-oss-20b
Hugging Face Inference Providers

The application is deployed as a live Streamlit web app.

Why Does Open Innovation Matter?

Open innovation made it possible for me to experiment with an open-weight AI model and build a useful application without depending entirely on a closed AI API.

Using openai/gpt-oss-20b through Hugging Face gave me the flexibility to integrate an open-weight model into my own application and customize the AI's behavior for language-learning practice.

For this project, open AI makes experimentation and learning more accessible to developers who want to understand and build with modern AI systems.

My Agent Session

I did not use a DevRelay agent session for this project.

Prize Categories

  • AI / Open Innovation
  • Education / Language Learning
  • Open Source

Top comments (0)