Software Engineer & Builder, Animator, Structural Engineer, Educator
I design and build digital products that are fast, accessible, and crafted with attention to detail. Focused on full-stack development with a strong interest in developer tooling and open-source.
Built and maintained multiple client-facing applications. Introduced CI/CD pipelines that cut deployment time by 60%. Worked closely with design to ship pixel-perfect interfaces.
Started career in high school, learnt computer science for four years that was capped with a final year project. Enrolled into an online Python bootcamp in Udemy where I undertook rigourous training in backend and frontend development. Learnt the full software delivery lifecycle end-to-end by spending years building and shipping projects, both personal and for clients. Participated in hackathons on campus and won a few. Also contributed to open source projects and built some of my own.
An android application that enables users to make M-pesa transactions by tapping their phones together or against POS systems. The app utilises the NFC capabilities of modern smartphones to facilitate secure and convenient peer-to-peer payments, as well as payments to merchants. This is made possible through the use of a double rail architecture, where the app detects whether the user has an active internet connection. If the user is online, the app processes transactions in real-time through the M-pesa API. If the user is offline, the app utilises GSM technology to make transactions by sending USSD codes directly to the M-pesa system. This ensures that users can continue to make payments and manage their finances regardless of their connectivity status, providing a reliable and flexible payment solution for a wide range of users. Tapesa transforms a phone into a contactless payment device.
A open source system that integrates into any AI model in the market to provide real-time data to the model. It uses a combination of web scraping, APIs and databases to collect and store data on various topics such as news, weather, stock prices and more. Chronos then provides this data to the AI model in a structured format, allowing it to make informed decisions and generate more accurate responses based on the most up-to-date information available.
An Nba betting prediction system that uses different features such as player stats, team stats, injuries and more to make predictions on the outcome of NBA games. The data scraped from the nba api is organized and used to train machine learning models such as XGBoost and Random Forest to make predictions on the outcome of NBA games. The system also uses a custom implementation of the GNL algorithm to optimize the betting strategy and maximize profits. The system backend is complemented with a dashboard for user friendliness.
A platform that provides historical and real-time farm performance analysis for Kenyan tea farmers, based on real data from KTDA. Farmers can keep track of their farm's performance, compare with other farms, and get insights to improve yield and quality. The system uses machine learning models, XGBoost, SARIMA and GNL to make predictions on monthly yields, monthly prices and yearly bonuses based on historical data. Also field officers can use the platform to monitor the performance of the farms under their supervision and provide targeted support to farmers who are struggling. The platform is built with React.js for the frontend, flask for the backend and MongoDb for the database. It also integrates with external APIs to get real-time weather data and market prices.
Languages
Frontend
Data Science
Backend
Infrastructure
Databases
Structural Engineering
Animation
Other
Whether you have a project in mind, want to collaborate on something open-source, or just want to say hi — my inbox is always open.
Send me an email ↗