During my internship at AI Guru, Ahmedabad, I worked on hands-on AI and Machine Learning applications that go beyond tutorials. These projects reflect real problem-solving in computer vision — from building object detection pipelines to training custom deep learning models.
The highlight is the use of YOLOv8 (You Only Look Once, version 8) — a state-of-the-art real-time object detection model widely used in industry and research due to its speed and accuracy.
This repository contains four core projects built during my internship at AI Guru, Ahmedabad — each representing a different stage of my learning curve.
A complete hands-on workflow for training custom object detection models using YOLO.
📓 Notebook:
https://github.com/Payal-mak/AI-Guru/blob/main/Train_YOLO_Models.ipynb
What this covers:
- Understanding YOLO architecture and workflow
- Preparing datasets in YOLO format
- Training custom models
- Evaluating performance
- Running inference on new images
This project pushed me from using models to actually building and training them. It taught me how data, configuration, and compute come together to create intelligence.
A data-driven dashboard built using Streamlit, powered by real census data from the official Census of India website.
Highlights:
- Uses authentic government datasets
- Interactive visualizations for population trends
- Filters by state, year, and demographic parameters
- Focused on turning raw data into meaningful insight
This project taught me how real data behaves — messy, inconsistent, and full of edge cases — and how to design systems that still make sense of it.
My first React project, built to understand how modern frontend frameworks actually work.
What I explored:
- Component-based architecture
- State management
- Event handling
- Dynamic UI updates
Simple on the surface, but foundational in shaping how I think about frontend logic and user interaction.
A responsive web app that fetches and displays real-time weather data.
🛠️ Tech Stack
- React JS
- Vite
- OpenWeatherMap API
- CSS for styling
Features:
- Search by city
- Live temperature and conditions
- Clean, minimal UI
- API integration and error handling
This project pushed me to think beyond static pages and build applications that communicate with the real world.
✔ Python — scripting and model implementation
✔ YOLOv8 — object detection deep learning model
✔ Ultralytics — Python library for YOLO models
✔ OpenCV — image/video processing
✔ Jupyter Notebooks — experiment tracking and demonstration
✔ Git & GitHub — version control and collaboration