dev.gxuri.in

Fri Jul 25 2025

NoCode AI Model Builder

NoCode AI Model Builder

Next.js

FastAPI

Machine Learning

Tailwind CSS

A no-code platform for building and training machine learning models with an intuitive drag-and-drop interface, powered by Next.js and FastAPI.

What is NoCode AI Model Builder?


Introduction

NoCode AI Model Builder is a user-friendly platform designed to democratize machine learning by enabling users to build, train, and evaluate AI models without coding expertise. Built with Next.js for a responsive frontend and FastAPI for robust backend processing, it supports automated data preprocessing, feature engineering, and model training. The platform integrates popular ML algorithms like Decision Trees, KNN, Linear Regression, Random Forest, and XGBoost, with hyperparameter tuning via GridSearchCV.

Features

  1. Intuitive Drag-and-Drop Interface: Create ML workflows using a visual flowchart interface powered by React Flow.
  2. Automated Data Preprocessing: Handles missing values, outliers, and feature encoding with intelligent strategies (e.g., StandardScaler, OneHotEncoder).
  3. Model Training and Evaluation: Supports multiple regression models with automated hyperparameter tuning and metrics like MSE, R², and MAE.
  4. Data Visualization: Generates histograms, bar plots, and correlation heatmaps for exploratory data analysis.
  5. File Support: Processes CSV, Excel, and JSON datasets with dynamic handling of numeric and categorical data.
  6. Robust Backend: FastAPI backend ensures efficient data processing and model training with scalability.
  7. Responsive UI: Built with Tailwind CSS, Shadcn UI, and Recharts for a modern, user-friendly experience.

User Workflow

  1. Users upload a dataset (CSV, Excel, or JSON) via the frontend interface.
  2. The platform preprocesses the data, handling missing values, outliers, and encoding categorical features.
  3. Users select a target column and train models using a drag-and-drop interface.
  4. The system automatically tunes hyperparameters and evaluates models with metrics like MSE and R².
  5. Users can view visualizations and download processed datasets or trained models.
  6. Results are displayed in a responsive UI with summary statistics and charts.

Tech Stack

  • Next.js ^15.1.6: React-based framework for the frontend.
  • FastAPI: Python-based backend for data processing and model training.
  • Scikit-learn: Machine learning library for model training and evaluation.
  • Tailwind CSS ^3.4.1: Styling framework for a responsive UI.
  • Shadcn UI: Component library for consistent UI elements.
  • Recharts ^2.15.1: Data visualization library for charts.
  • React Flow ^11.11.4: Library for building interactive flowcharts.

TechXelerate 2025 Hackathon

This project was developed during the TechXelerate 2025 Hackathon held at BITS Pilani, Goa Campus. Our team secured 2nd place in our domain, showcasing the platform's potential to simplify machine learning for non-coders. Below is a photo from the event:


Get Involved

We welcome contributions to enhance the NoCode AI Model Builder platform. You can contribute by:

  • Improving model support or adding new algorithms.
  • Enhancing visualization options or UI components.
  • Optimizing backend performance for larger datasets.

Visit our GitHub repository to explore the codebase, report issues, or submit pull requests.

Stay Connected

Follow the project for updates and announcements: