Mzuzu, Malawi · Mzuzu University
Gomezgani.
Data Scientist & Systems Builder
Data Science graduate focused on Machine Learning and Generative AI engineering, with hands-on experience building end-to-end intelligent systems from data pipelines and model development to APIs and production applications. Experienced across Python, machine learning, backend engineering, and modern AI architectures including RAG and LLM-powered applications. Building reliable, data-driven AI systems for real-world problems.
Selected Work
Featured Projects
Real systems, real impact. These projects address specific, niche problems, built to production standards.
Data Science
SmsSava
SMS-Based Personal Finance Tracker for Unbanked Malawians
Parses raw Airtel/TNM mobile money SMS alerts to auto-categorize spending and build personal financial profiles — no app or internet required.
Data Science / ML
CreditIQ
ML-Powered Credit Scoring Engine for Informal Economy Workers
A machine learning system that builds credit scores from non-traditional signals — mobile money history, airtime top-ups, and social utility patterns — for people without formal credit histories.
Agri-tech / Data Science
AgriPulse
Crop Yield Prediction & Advisory System for Smallholder Farmers
Predicts maize and soybean crop yields in Malawi using rainfall, soil type, and historical FAOSTAT data, with a Streamlit advisory dashboard for extension workers.
7
Projects Built
4+
ML Systems Shipped
BSc
Data Science
MW
Based in Malawi
Who I Am
Building from Malawi,
thinking for the world.
I'm a Data Science graduate building ML systems and AI-powered applications at Mzuzu University. I don't just train models: I build the pipelines, deploy the APIs, and ship the application. End-to-end is the only way I work.
My work focuses on real problems: household nutrition risk in Malawi, financial inclusion via SMS, blockchain trust systems, and network security. Each project is built with production engineering in mind.
Core Disciplines
Capabilities
AI Engineering
From retrieval pipelines to production deployment, covering the full stack of modern AI system development.
RAG
Retrieval pipelines, chunking strategies, embedding models, vector search, reranking, and grounded generation.
→ NutriFarm AIAI Agents
Tool-calling, workflow orchestration, state management, memory, and human-in-the-loop execution.
Evaluation
Retrieval evaluation, hallucination testing, model performance metrics, and system-level benchmarking.
→ NutriFarm: 97.4% accuracyProduction AI
API integration, authentication, observability, containerisation, and cloud deployment.