Loading portfolio...
Loading portfolio...
Hi, I’m Isaac Mungai. My journey in tech started with curiosity—how do systems work, and how can we make them smarter? That curiosity grew into a passion for machine learning and software engineering, where I now spend most of my time building, experimenting and refining ideas through code.
As a computer science major, I’ve had the chance to explore everything from designing software systems to developing AI-driven applications. Hackathons, side projects and collaborative work have pushed me to think faster, adapt and turn abstract ideas into real solutions. What drives me is the challenge of solving problems that matter and the thrill of knowing there’s always something new to learn, test and build.

Crop Doctor is an intelligent assistant for farmers, delivered through an intuitive mobile app. It empowers farmers to identify and resolve crop issues quickly and effectively. By simply uploading a photo of a crop or describing its symptoms in the app, a farmer can get a reliable diagnosis and a clear action plan.
GitHub Activity Dashboard – An interactive web app that visualizes my GitHub activity, including repositories, commits, and contribution trends. Built with Next.js, React, Tailwind CSS, and ShadCN UI, it offers real-time stats, search & filtering, and responsive charts to showcase my coding journey.
Digit Recognizer — Handwritten Digit Classification with KNN Developed a machine learning model using the K-Nearest Neighbors (KNN) algorithm to classify handwritten digits from the MNIST dataset, achieving an accuracy of 91%. Implemented data preprocessing and model tuning to optimize performance. Deployed the application with an interactive Streamlit interface for real-time digit input and prediction.
SugarCheckAI — Diabetes Risk Prediction Tool Developed and deployed a machine learning web app that predicts the likelihood of diabetes diagnosis based on user-provided health metrics. Utilized a Random Forest Classifier combined with SMOTE to handle class imbalance and improve model performance. Deployed the model using Streamlit for a user-friendly interface.
KAFU
Oracle
edX
AWS
AWS
Alternatively, reach me via email or on my socials.