VISHWAS DOJO

A sacred space for mastering the art of code, training the discipline of innovation, and perfecting the way of technology

Craftsmanship
Intelligence
Precision

The Way of the Dojo

道場の道

In the traditional dojo, mastery comes through relentless practice, deep understanding, and the pursuit of perfection. Here, code becomes kata, algorithms become meditation, and every project is a step along the path of technological enlightenment.

From machine learning architectures to full-stack applications, each creation embodies the discipline and precision that defines the martial arts mindset applied to software craft.

Live Demonstrations

Interactive projects showcasing advanced AI and computer vision capabilities

TaskAI

Live Project

An advanced productivity analytics platform that leverages natural language processing and machine learning to transform unstructured task descriptions into hierarchical, actionable workflows. The system performs real-time semantic analysis on user input to extract implicit task attributes including priority levels, urgency indicators, and contextual dependencies—eliminating manual categorization overhead.

Flask PostgreSQL OpenAI NGINX Gunicorn Jinja SQLAlchemy

RAG AI Teacher Interface

Live Project

Advanced conversational AI with Retrieval-Augmented Generation capabilities, featuring real-time responses and document understanding.

React TypeScript RAG OpenAI Vector DB

Fruit Freshness Detector

Live Project

Computer vision application that analyzes fruit images to determine freshness using deep learning models.

Python TensorFlow OpenCV Computer Vision CNN

Anime style Journal

Live Project

Computer vision application that analyzes fruit images to determine freshness using deep learning models.

Python TensorFlow OpenCV Computer Vision CNN Midjourney

Portfolio Showcase

Core projects demonstrating machine learning expertise and analytical thinking

California House Regression

Developed a custom feature engineering pipeline to extract complex geospatial patterns from raw coordinates. By implementing K-Means clustering to identify hotspots and applying weighted RBF kernels for proximity modeling, the solution achieved a quantifiable improvement in predictive scores.

Python Pandas Scikit-learn Random Forest Linear Regression Grid Search Random Search

Time Series Forecasting Suite

Comprehensive forecasting toolkit implementing ARIMA, LSTM, and Prophet models for financial and business analytics.

Python Prophet TensorFlow Time Series Finance

Neural Network from Scratch

Educational implementation of multilayer perceptron with backpropagation, showcasing fundamental deep learning concepts.

Python NumPy Deep Learning Neural Networks

Connect & Collaborate

Ready to embark on a journey of innovation? Let's build something extraordinary together.