CoderCoach

Inspiration My group and I were inspired by the various coding problems we've experienced in school and wondered about the underlying patterns behind them. What it does The agent takes input in the form of a JSON file (with patterns) and suggests hints to solve the problem based on associated pattern structure and "tells". How we built it This project was built using the React framework with TypeScript (frontend) and Python + Gemini API (backend). Challenges we ran into One of the biggest challenges we ran into was integrating the frontend and backend connectivity, as well as shaping the UI to fit our needs. Accomplishments that we're proud of We were able to successfully call the API to give responses to our demo patterns.json file. What we learned There are no limits to the human capacity to produce efficient solutions to even the most mundane of problems. What's next for CoderCoach Potentially adding internal validation checks to make output more human-centered.

Built With

Submitted to

HackRU Fall 2026

Created by

I contributed to the backend development of CoderCoach, an AI-powered coding assistant designed to help students understand programming concepts instead of simply giving them answers. I worked on integrating the Gemini API, managing API keys and environment variables, and developing the logic for detecting common coding patterns. I also worked on processing AI responses, validating detected patterns, and retrieving relevant hints and practice questions. My focus was on helping the application guide students through problems step by step while encouraging them to develop their own problem-solving skills.

Abhigya Sinha
I helped make the coding patterns for our project. I added hints and practice problems using business examples like sales, budgets, and invoices, then saved them in JSON file and shared it on GitHub. This was my first hackathon project, so I was learning as I went. I learned how to use VS Code, write JSON, and save and sync changes with Git.

Yogishi751
Wisdom Oyemi