AI & Full-Stack Engineer | 5⭐ CodeChef | AI Agents, Automation, Backend Systems | Ex-Microsoft AI & EY GDS

gorav jindal  • Peerlist

➡️Developed "Chat to PDF," a web application enabling users to interact with PDF documents through natural language conversations using Google's Gemini API. ➡️Engineered the application to parse and extract text from PDF documents, creating an index that allows for efficient information retrieval. ➡️Implemented a chat interface that utilizes Gemini's natural language processing to understand user questions and provide contextually relevant answers based on the PDF content. ➡️Integrated Gemini API to enable users to ask any question related to the PDF, regardless of the complexity or phrasing. ➡️Utilized Python, PDF parsing libraries, Gemini API, and React.js to create a seamless and intuitive user experience. ➡️Addressed the challenge of accessing and understanding information within large PDF documents by providing an AI-driven conversational interface. ➡️Demonstrated proficiency in AI integration, PDF parsing, and web development through the successful creation of "Chat to PDF." ➡️Improved user productivity by enabling quick and accurate retrieval of information from PDF documents through conversational interaction.

AI-Powered PDF Interaction

This is an MCP (Model Context Protocol) server that executes Manim animation code and returns the generated video. It allows users to send Manim scripts and receive the rendered animation.

Animation Video MCP

BlenderMCP connects Blender to Claude AI and cursor through the Model Context Protocol (MCP), allowing Claude and cursor to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.

Blender MCP

➡️Developed a predictive machine learning model to accurately forecast the likelihood of diabetes in patients based on relevant health indicators. ➡️Engineered the model using [mention specific algorithms, e.g., Logistic Regression, Random Forest, Support Vector Machines, Neural Networks] to analyze and interpret patient data. ➡️Implemented data preprocessing techniques, including data cleaning, feature scaling, and feature selection, to optimize model performance and accuracy. ➡️Utilized [mention specific libraries, e.g., Python's scikit-learn, TensorFlow, or PyTorch] to build, train, and evaluate the predictive model. ➡️Trained the model on [mention data source, e.g., Pima Indians Diabetes Database] to ensure robust and reliable predictions. ➡️Evaluated model performance using metrics such as accuracy, precision, recall, and F1-score to validate its effectiveness. ➡️Addressed the challenge of early diabetes detection by providing a data-driven tool for risk assessment. ➡️Demonstrated proficiency in machine learning, data analysis, and predictive modeling through the development of this diabetes prediction model. ➡️Improved potential patient outcomes by enabling early detection and preventative measures for diabetes.

Diabetes Prediction

I made this game using python and pygame library. The game story is as follows: Big problem! The holidays are almost here but people keep trying to steal your presents. It's up to you to see how long you can defend your precious gift. The source code of project is - https://github.com/Gorav22/Dont-touch-presents

Dont Touch My Presents

- Capture your real time photo - Can tell emotions of the person like sad, happy or angry. - Can be installed like an app or can work like an website. - Accuracy of emotions predicting is 95% - Used nodejs, react and face-api with face-api models

Face Recognition

Developed "Ghumna," an AI-driven travel planning web application designed to create personalized travel itineraries based on user preferences. Engineered the application to leverage AI algorithms for destination recommendations, itinerary generation, and activity planning. Implemented a natural language processing (NLP) interface that allows users to input their travel preferences, budget, and desired experiences in conversational language. Integrated with travel APIs [mention specific APIs, e.g., Google Maps API, travel booking APIs, destination information APIs] to retrieve real-time data on flights, accommodations, and attractions. Utilized [mention specific technologies, e.g., Python for backend, NLP libraries, React.js for frontend] to create a user-friendly and efficient travel planning experience. Addressed the challenge of time-consuming travel planning by providing an automated and personalized itinerary generation tool. Demonstrated proficiency in AI integration, NLP, and web development through the creation of "Ghumna." Enhanced user travel experiences by providing tailored itinerary plans that optimize travel time and maximize enjoyment.

Ghumna

➡️Developed "Growthmock," an AI-powered web application designed to simulate realistic mock interviews and provide comprehensive feedback. ➡️Engineered the platform to generate personalized mock interview scenarios and questions using advanced AI algorithms, tailored to specific job roles and industries. ➡️Implemented a real-time feedback system that analyzes candidate responses, providing insights into communication skills, technical proficiency, and overall interview performance. ➡️Integrated natural language processing (NLP) to evaluate answer quality and provide constructive feedback on areas for improvement. ➡️Created a feature that generates targeted preparation questions, enabling users to focus their study efforts on key interview topics. ➡️Utilized React.js, Node.js, and Python with NLP libraries to build a robust and user-friendly platform. ➡️Addressed the challenge of interview preparation by providing an accessible and personalized AI-driven solution for practicing and improving interview skills. ➡️Demonstrated proficiency in web development, AI integration, and user experience design through the development of "Growthmock." ➡️Addressed the common problem of interview anxiety by creating a safe space to practice and receive feedback.

Growthmock

I developed an innovative Python tool that enables users to control their laptop or PC without using a touchpad or keyboard. Leveraging the power of OpenCV and MediaPipe, this project transforms hand gestures into actionable commands, providing a seamless and intuitive user experience. Key Features: Gesture Recognition: Utilizes MediaPipe for real-time hand tracking and gesture detection. Customizable Commands: Allows users to map specific gestures to various system actions, such as scrolling, clicking, and navigating between applications. Enhanced Accessibility: Ideal for users with mobility impairments, offering an alternative to traditional input methods. User-Friendly Interface: Designed with a focus on ease of use, ensuring that anyone can set up and start using the tool with minimal effort. Technologies Used: - Python: The core programming language for developing the tool. - OpenCV: For image processing and computer vision tasks. - MediaPipe: For robust hand tracking and gesture recognition. Impact: This project not only showcases my proficiency in computer vision and machine learning but also demonstrates my commitment to creating accessible and user-centric technologies. By eliminating the need for physical input devices, this tool opens up new possibilities for interaction and control, making technology more inclusive and adaptable to diverse user needs.

Handler- Gesture-Controlled PC Interface

➡️Developed "Diet Analyzer," a web application that estimates the caloric content of food items in uploaded images using image recognition and AI. ➡️Engineered the application to process uploaded images, identify individual food components, and estimate their respective calorie values. ➡️Implemented image recognition models using TensorFlow and image classification APIs to accurately identify food items within the uploaded image. ➡️Integrated a calorie database or API to correlate identified food items with their nutritional information and calculate calorie estimates. ➡️Utilized Python, image processing libraries, and React.js to build a user-friendly and efficient calorie estimation tool. ➡️Addressed the challenge of manual calorie tracking by providing an automated and image-based approach to dietary analysis. ➡️Demonstrated proficiency in image recognition, AI integration, and web development through the creation of "Diet Analyzer." ➡️Empowered users to make informed dietary choices by providing quick and accurate calorie estimations from food images.

Image-Based Calorie Estimation

I made a chatbot named as Jarvis AI. Here i use 2 api keys - Google Gemini Api Key (for text generation) - Weather api key for come to know about the humidity of any location. the source code is here https://github.com/Gorav22/Jarvisai

Jarvis Ai

What if your phone could control your entire laptop using plain English? JLaw is an AI agent that lives inside your laptop and takes orders from your phone — no screen sharing, no remote desktop, no lag. Just type or speak a command on your phone: → "Open Chrome and search for flights to Delhi" → "Move all PDFs from Downloads to Documents" → "Run my Python script" ...and your laptop does it. Built with Python, MCP Protocol, WebSockets, and REST APIs. This is what AI agents should feel like — invisible, fast, and actually useful.

JLaw — AI Employee Inside the Laptop

➡️Developed "gorav-readme-generator," an npm package that automates the generation of comprehensive README.md files using Google's Gemini AI. ➡️Engineered the npm package to analyze source code across multiple files, leveraging Gemini's natural language processing to produce accurate and informative project documentation. ➡️Implemented functionality to extract key code information, including file descriptions, function signatures, and variable definitions, and convert them into human-readable README content within the npm package. ➡️Utilized Node.js and [mention specific npm libraries used, e.g., fs, axios (if used for API calls to Gemini)] to enable seamless code analysis and AI-powered README generation as an npm package. ➡️Addressed the challenge of streamlining software project documentation for JavaScript and Node.js projects by providing an automated README generation tool as an easily installable npm package. ➡️Demonstrated proficiency in npm package development, AI integration, and code parsing through the successful creation and deployment of "gorav-readme-generator." ➡️Published "gorav-readme-generator" on npm, making it accessible to developers for simplified project documentation.

Readme generator Npm Package

SkillSniper is an innovative Chrome Extension designed to optimize your job search experience on LinkedIn. Leveraging advanced natural language processing with Ollama's Gemm3 model, SkillSniper analyzes job postings in real-time to provide personalized insights and recommendations. This powerful tool helps you determine if a job is the right fit for your skills and experience, and offers actionable advice on how to enhance your resume and LinkedIn profile to better match job requirements. Key Features: Intelligent Job Analysis: Utilizes Ollama's Gemm3 model to analyze job postings and provide instant feedback on job fit. Resume and Profile Optimization: Integrates PyPDF to extract and analyze your resume, offering tailored recommendations to improve your application. Seamless Integration: Built with Uvicorn for robust performance, ensuring a smooth and responsive user experience directly on LinkedIn. Real-Time Insights: Automatically analyzes job postings and provides feedback on the page, helping you make informed decisions quickly. Benefits: Save Time: Quickly identify jobs that match your profile and focus on applying to the right opportunities. Enhance Your Application: Get data-driven insights to improve your resume and profile, increasing your visibility to recruiters. Stay Competitive: Gain an edge in the job market with personalized recommendations and a streamlined job search process. SkillSniper is your ultimate career companion, helping you navigate the job market with confidence and efficiency. Try it today and take a significant step towards a successful career journey!

SkillSniper: Revolutionize Your Job Search on

Terminal MCP Server is a Model Context Protocol (MCP) server that allows executing commands on local or remote hosts. It provides a simple yet powerful interface for AI models and other applications to execute system commands, either on the local machine or on remote hosts via SSH.

TerminusAI

🚨 Introducing My Latest Project: AI-Powered Online Exam Proctoring System 🚨 📍 Built with Computer Vision + AI to uphold exam integrity in the digital era. With the rise of remote learning, maintaining the fairness of online exams has become more challenging than ever. That’s why I built an intelligent Online Exam Proctor System — designed to detect suspicious behavior and ensure a secure test environment. 🧠💻 🔐 Key Features: 🔹 Real-time face verification and liveness detection 🔹 Movement tracking to prevent distractions or impersonation 🔹 Screen monitoring for tab switching and unauthorized key use like copy paste commands detection. 🔹 Voice activity analysis to detect surrounding conversations 🔹 Admin dashboard for live monitoring, violation records, and student reports 🛠️ Tech Stack: Python | OpenCV | Flask | YOLOv8 | Dlib | MediaPipe | Oracle MySQL  | PyAutoGUI | PyGetWindow 💻 Try it out or explore the code: 👉 GitHub Repository - https://lnkd.in/gt9gR_Ai 👉 Demo video link- https://lnkd.in/ga_VPvCU 🔁 Feel free to connect if you're working on something similar or want to collaborate!

The-Online-Exam-Proctor

this is zoom clone where u can use features are -videocalling -zoom link for join -join by room no. -chat option To see the project search https://github.com/Gorav22/zoom-clone