Inspiration
Learning to code often feels passive and isolating. Many learners struggle not because of lack of ability, but due to missing real-time feedback, motivation, and practical pressure. Competitive coding platforms can be intimidating, while traditional learning platforms lack interaction.
We wanted to create a platform that makes coding engaging, competitive, and guided, combining the excitement of real-time challenges with AI-powered mentorship. This idea led to Slashcoder.
What it does
Slashcoder is an AI-powered competitive coding and learning platform where developers practice coding through real-time 1v1 battles, challenges, and guided problem-solving. It provides instant code evaluation, live competition, leaderboards, and Gemini-powered AI feedback that helps users understand mistakes and improve logically rather than memorizing solutions.
How we built it
Slashcoder is designed as a cloud-native application on Google Cloud: Frontend: React with Monaco Editor for a real-time coding experience Backend: Node.js with WebSockets for live match synchronization Authentication: Firebase Authentication Database: Firestore for users, matches, and rankings AI: Gemini is used to analyze code, explain errors, generate hints, and personalize learning Deployment: Built to scale using Google Cloud–compatible infrastructure This architecture enables low-latency interactions and scalable real-time battles.
Challenges we ran into
Real-time synchronization between players during live matches Designing AI prompts that guide learners without revealing solutions Preventing misuse of AI during competitive battles Optimizing performance for mobile and low-bandwidth users Each challenge required careful engineering and iteration.
Accomplishments that we're proud of
Successfully built a real-time coding battle system Integrated Gemini AI to provide meaningful learning feedback Designed a scalable, cloud-native architecture Created an engaging platform that turns coding practice into an interactive experience Achieved early user adoption and positive engagement feedback
What we learned
AI is most effective as a mentor, not a replacement for problem-solving Real-time systems require robust state management and failure handling Cloud-native tools significantly simplify scalability and reliability Engagement increases when learning is interactive and competitive Using Gemini and Google Cloud taught us how advanced AI and scalable infrastructure can transform education.
What's next for Slashcoder
Team battles and tournaments AI-generated and adaptive coding challenges Advanced learning analytics using Gemini Recruiter-ready skill profiles and assessments Deeper integration with Google Cloud services for scale and reliability
Built With
- and-learning-feedback-authentication:-firebase-authentication-database:-cloud-firestore-cloud-platform:-google-cloud-platform-(gcp)-hosting-/-deployment:-cloud-native-deployment-(cloud-run?ready)-version-control:-git
- chatgpt
- cloud-firestore
- express-real-time-communication:-websockets-(socket.io)-ai-/-llm:-gemini-(google-ai)-for-code-analysis
- express.js
- firebase-authentication
- gemini
- git
- github
- google-cloud
- hints
- javascript
- judge0
- monaco-editor-backend:-node.js
- monacoeditor
- node.js
- openai
- python
- react
- typescript-frontend:-react
- websockets
Log in or sign up for Devpost to join the conversation.