📍 Gaborone, Botswana  •  🎓 MEng Biomedical Engineering, UCL  •  💼 Orange Botswana

• 🚌 TselaGoo — A combi (minibus taxi) route-finder for Gaborone. Built with FastAPI + Supabase on the backend and React 19 + TypeScript + Vite + Tailwind on the frontend, with Google Maps JS API for rendering. Features include per-segment route geometry storage, OSRM-based polyline computation from GPX tracks, two-leg route visualisation with transfer pins, and haversine-based nearest-stop search.

• 📚 ExamAI — An AI-powered exam practice platform for BGCSE and JCE students in Botswana. Built with FastAPI, RAG (sentence-transformers + OpenAI), and a custom PDF parser to ingest past papers. Generates contextually grounded practice questions and explanations from real exam content.

• 🤖 AI-Chatbot — A web-based conversational AI assistant built with Python (Flask) and OpenAI's GPT-3.5 / GPT-4 API. Real-time chat UI via JavaScript and Flask, secure key management with .env, and a modular architecture ready for deployment on Render, Replit, or Hugging Face Spaces.

• ❤️ Heart Failure Prediction Dashboard — A clinical-decision-support dashboard that estimates heart failure risk from patient features. K-Nearest Neighbors classifier (with cross-validated k selection and Manhattan distance) trained on heart failure clinical records, served via a Dash UI with scikit-learn, StandardScaler preprocessing, and a live confusion matrix rendered with seaborn + matplotlib.

• 🏠 House Price Prediction Dashboard — A machine learning web app that predicts house prices from user input. Random Forest Regressor trained on the Ames Housing dataset, served through an interactive Dash + Bootstrap dashboard. Demonstrates end-to-end ML deployment from preprocessing to live prediction.

• 🧠 Machine Learning Coursework (MPHY0041) — UCL biomedical ML coursework covering classification techniques for dementia diagnosis (CN vs MCI vs AD). Includes manual implementations of Iteratively Reweighted Least Squares (IWLS), gradient descent, and L2-regularised logistic regression, with comparison against sklearn.

• 🔬 4th-Year UCL Project — Final-year biomedical engineering project work. Arduino-based instrumentation for sensor sensitivity testing and calibration.

• 📊 Daily Attendance Analyser — A Python data-analysis script that processes CSV attendance datasets, computes Present/Absent percentages, and generates comparative bar charts using pandas and matplotlib.

I'm a Biomedical Engineer with a strong software-engineering tilt — currently focused on shipping practical, real-world tools for the Botswana market while continuing to work across medical imaging, machine learning, and full-stack web development.

Day to day I work at Orange Botswana, and outside of that I'm building TselaGoo. I care about software that's actually used — clean architecture, fast feedback loops, and shipping over polishing.

• 🔭 Currently building TselaGoo (Gaborone combi route finder)
• 🌱 Currently learning advanced ML deployment patterns, scalable backend design, and real-time GPS tracking
• 💬 Ask me about FastAPI, Supabase, React, medical imaging, or building for emerging markets
• 📫 Reach me through GitHub

Languages

Frameworks & Libraries

Databases & Cloud

Tools

University College London (UCL)  ·  London, UK Integrated Master of Engineering — Biomedical Engineering

• Specialisations in medical imaging, mathematical modelling, machine learning, and software engineering
• Project work in pelvis MRI segmentation using supervised and semi-supervised deep learning on T2-weighted scans
• Coursework spanning Python, MATLAB, C++, and full-stack development

Orange Botswana  ·  Gaborone, Botswana Working within IT & Digital Transformation.

• 🇧🇼 Building TselaGoo to make Gaborone's combi network easier for everyone to navigate
• 🧬 Background spans medical imaging, deep learning, and full-stack web — I like projects that sit on the line between hardware, data, and product
• ♟️ Outside of code: investing, sports, and exploring how tech can solve everyday problems in Botswana