From infrastructure to intelligence — built to ship both.

I'm an ML engineer who started in cloud infrastructure. My goal is to build machine learning systems that aren't just accurate in notebooks — but precise, observable, and production-ready.

The Journey

I started with a degree in Information Technology and a fascination for how systems hold together under load. That led me to dual AWS certifications as a Solutions Architect and Cloud Practitioner — I didn't just want to build the pipes; I wanted to understand what was flowing through them.

That curiosity pulled me into data science through the ALX Africa x Explore AI Academy program, where statistical modeling and machine learning replaced infrastructure diagrams on my whiteboard. Then a production incident sharpened everything.

Auditing an M-Pesa churn model, I found a 36-point precision gap between reported training accuracy (87%) and real-world performance — causing 66% false positives and a 91% campaign failure rate. That gap is what I now engineer against.

Today I'm a data science mentee at Phoenix Analytics Africa, collaborating with a team of analysts on predictive models, data pipelines, and deployment challenges — while independently deepening my production ML engineering skills across FastAPI, Docker, and MLOps.

ML & Modeling

  • • Scikit-Learn, TensorFlow/Keras, XGBoost
  • • Feature engineering & EDA
  • • Model evaluation & precision auditing
  • • Statistical modeling & data preprocessing

Engineering & Deployment

  • • FastAPI & RESTful API design
  • • Docker & Kubernetes
  • • CI/CD (Travis CI, GitHub Actions)
  • • Django & Django REST Framework

Cloud & Infrastructure

  • • AWS Certified Solutions Architect
  • • AWS Certified Cloud Practitioner
  • • Linux systems & scalable architecture
  • • PostgreSQL & MySQL

Data & Analytics

  • • Python — Pandas, NumPy, SQL
  • • Power BI & data visualization
  • • Pipeline design & data quality
  • • Go, C++

What I'm Doing Now

Right now I'm building a YouTube trending video analysis pipeline — a Kenya-vs-global comparison using PostgreSQL, Python, and a GPU-accelerated translation pipeline — while continuing mentorship work focused on forecasting, fraud detection, and recommendation systems.

I'm open to roles as a Machine Learning Engineer or Python / Data Engineer — particularly where production reliability, model observability, and measurable impact matter.

Let's Collaborate See the Code Download CV