Na it's fine

github

Debasish Tripathy • Peerlist

Code Pro is an open-source AI coding assistant and technical interview copilot designed for developers, students, and software engineers preparing for or participating in coding interviews.Code Pro provides real-time assistance during technical interviews by combining screen capture, audio transcription, AI-powered problem solving, code generation, reasoning, and interactive follow-up questions in a single desktop application.It is built as an open-source alternative to closed-source AI interview tools such as Cluely, Interview Coder, and Final Round AI, giving developers more control over their data, AI provider, prompts, costs, and customization.🚀 Features<h3>👁️ Screen Capture &amp; Problem Understanding</h3>Code Pro can capture coding problems directly from your screen and use AI vision capabilities to understand:<ul><li>Coding interview questions</li><li>Problem statements</li><li>Constraints</li><li>Input and output formats</li><li>Examples</li><li>Test cases</li><li>Existing code</li><li>Error messages</li><li>Interviewer-provided coding tasks</li></ul>Instead of manually copying an interview question into an AI chatbot, Code Pro can extract the relevant context directly from your screen.<h3>🎧 Real-Time Audio Transcription</h3>Code Pro supports separate audio sources for better interview context:<ul><li>System audio — captures and transcribes the interviewer</li><li>Microphone audio — captures what you say</li><li>Separate transcription streams help distinguish interviewer questions from your own responses</li></ul>This allows the AI to maintain context throughout a technical interview and respond to changing questions or requirements.<h3>🧠 Reasoning-First AI Solutions</h3>Code Pro is designed to provide more than a blind code dump.Solutions can be structured around:<ol><li>Understanding the problem</li><li>Key observations</li><li>Intuition</li><li>Algorithm and approach</li><li>Complexity analysis</li><li>Step-by-step dry run</li><li>Implementation</li><li>Edge cases</li><li>Possible optimizations</li></ol>This makes the generated response useful for understanding and explaining a solution, rather than simply copying code. <h3>💬 Interactive Follow-Up Questions</h3>Technical interviews are interactive, so Code Pro supports follow-up questions and changing requirements.You can ask questions such as:<ul><li>Why use a heap instead of two pointers?</li><li>Can this solution be optimized?</li><li>Can we reduce the space complexity to O(1)?</li><li>What happens with duplicate values?</li><li>Why does this algorithm work?</li><li>What is the time complexity?</li><li>Can you explain this line of code?</li><li>What edge cases should I consider?</li><li>Can we solve this using dynamic programming?</li><li>Is there a better approach?</li></ul>The AI can use the existing problem and conversation context to answer without requiring you to start over.🔑 Bring Your Own Key (BYOK)Code Pro follows a Bring Your Own Key (BYOK) model.You can connect your own API keys from supported AI providers such as:<ul><li>Google Gemini</li><li>OpenAI</li><li>Anthropic Claude</li></ul>Your requests can go directly to the AI provider instead of being routed through a proprietary SaaS backend.This provides greater control over:<ul><li>AI models</li><li>API usage</li><li>Costs</li><li>Prompts</li><li>Privacy</li><li>Configuration</li></ul>There is no requirement to pay for another monthly AI subscription just to access the application.💰 Pay for AI Usage, Not an Expensive SubscriptionTraditional AI interview assistants can charge significant monthly subscription fees.With Code Pro, you can use your own API key and pay the AI provider directly for usage.For inexpensive models, an individual coding problem can cost only a few cents or less depending on the model, prompt size, image processing, audio transcription, and usage.This makes Code Pro particularly useful for developers who want an affordable Cluely alternative, Interview Coder alternative, or Final Round AI alternative without another recurring subscription. 🔓 Fully Open SourceCode Pro is designed to be hackable and customizable.The project is released under the AGPL-3.0 license, allowing developers to inspect the source code, modify the application, customize prompts, add functionality, and create their own versions in accordance with the license.You can customize things such as:<ul><li>AI prompts</li><li>Solution formatting</li><li>Supported AI providers</li><li>User interface</li><li>Keyboard shortcuts</li><li>Screen capture workflows</li><li>Audio processing</li><li>Interview workflows</li><li>Model selection</li><li>Response structure</li></ul>Instead of using a closed AI interview platform where the underlying implementation is hidden, Code Pro gives developers the ability to inspect and modify the software.🛠️ Technology StackCode Pro is built using modern desktop application technologies:<ul><li>Electron</li><li>React</li><li>TypeScript</li><li>AI vision models</li><li>Large language models</li><li>Speech-to-text</li><li>Screen capture</li><li>Audio processing</li></ul>The architecture is intended to make the application accessible to developers who want to contribute, fork the project, or build additional AI-powered developer tools.🎯 Use CasesCode Pro can be useful for:<ul><li>Software engineering interviews</li><li>Coding interviews</li><li>Technical interviews</li><li>LeetCode practice</li><li>Algorithm practice</li><li>Data structures practice</li><li>System design preparation</li><li>Competitive programming practice</li><li>Debugging</li><li>Learning algorithms</li><li>Understanding unfamiliar code</li><li>Practicing technical explanations</li><li>AI-assisted coding</li><li>Developer productivity </li></ul>

CODE-PRO

AI Food Pantry is an innovative platform designed to simplify meal planning and pantry management using AI technology. It allows users to track and organize their pantry items efficiently while providing personalized meal suggestions based on health preferences.

Key Features:
Pantry Management: Easily add and manage food items in your pantry.
Health Integration: Update health details to receive tailored meal recommendations.
Quick Actions: Utilize features like menu generation and item analysis to optimize your meal planning experience.
The platform is currently in beta, and users can access it for trial at https://foodpan.onrender.com. Your feedback will help us enhance the platform further!

AI Food Panetry

This project is a chatbot application that allows users to have conversations with fictional characters. It uses a combination of technologies to achieve this:

Flask: This is a Python web framework that provides the structure for the chatbot application. It handles things like routing user requests and serving responses.
Google GenerativeAI: This is an API from Google that allows you to generate text content based on prompts and instructions. In this case, it's used to create the character's responses to user questions.
Custom NLP (Natural Language Processing): This likely refers to additional code written by the developer to process user input and potentially tailor the character's responses further.
HTML/CSS/JavaScript: These are the building blocks for the web interface where users interact with the chatbot. HTML defines the structure of the page, CSS styles its appearance, and JavaScript adds dynamic functionality.
MongoDB: This is a NoSQL database that's used to store the token id, which might be related to user authentication or API access tokens.

Character AI

My team and I participated in the Jane Street Kaggle competition, where we developed a machine learning pipeline to forecast market trends using high-frequency financial data. Together, we implemented techniques like feature engineering, memory optimization, and ensemble modeling (LightGBM and CatBoost) to enhance prediction accuracy and model performance. We also ensured reliability through cross-validation and feature importance analysis. This collaborative project allowed us to apply our collective skills in data preprocessing and financial forecasting to tackle real-world challenges effectively.

Jane street real-time market data analysis

A modern web application built with React and TypeScript that provides AI-powered tarot card readings using Google's Gemini AI.

Overview
Mystic Tarot Reading offers an immersive digital tarot reading experience, combining traditional tarot wisdom with modern AI interpretation. Users can ask questions, receive personalized three-card readings, and engage in follow-up conversations about their readings.

Key Features
Interactive Card Selection: Draw three cards from a beautifully designed digital tarot deck
AI-Powered Readings: Get detailed interpretations of your cards using Google's Gemini AI
Personalized Experience: Ask specific questions and receive tailored readings
Chat Interface: Engage in follow-up discussions about your reading
Reading History: Save and revisit your past readings
Responsive Design: Seamless experience across desktop and mobile devices
Technical Stack
Frontend: React 18 with TypeScript
Styling: Tailwind CSS
AI Integration: Google Generative AI (Gemini)
State Management: React Hooks
Build Tool: Vite
Markdown Support: React Markdown with remark-gfm
Prerequisites
Node.js 18.x or higher
npm 9.x or higher
A Google Gemini API key
Getting Started
Clone the repository
Install dependencies with npm install
Add your Gemini API key to .env
Start the development server with npm run dev
For detailed setup instructions, see the Installation section below.

Mystic Tarot Reading

The AI Note Generator is an advanced AI-driven application designed to streamline the process of note-taking for meetings and classes. By leveraging state-of-the-art language models and retrieval-augmented generation (RAG), this system ensures high-quality, structured, and insightful notes.