Data Engineer & ML Practitioner (Anonymised)

is a results-driven Data Engineer and Machine Learning Practitioner based in Kampala, Uganda, with five years of deep expertise in designing and deploying data infrastructure, ETL pipelines, and production machine learning systems that drive tangible business outcomes for African enterprises. 's technical toolkit is both broad and deep. He is highly proficient in Python and its data science ecosystem — Pandas, NumPy, Scikit-learn, and Matplotlib — and has extensive hands-on experience with large-scale distributed data processing using Apache Spark and Apache Kafka. He has built and maintained production data pipelines with Apache Airflow, orchestrating complex DAGs that ingest, transform, and load terabytes of financial and telco data daily. His cloud data warehouse experience spans Google BigQuery and Snowflake, and he is skilled in designing efficient data models for both analytical and operational workloads. One of 's most significant professional achievements is the real-time fraud detection system he designed and deployed at MTN Uganda Digital, where he serves as Senior Data Engineer. This system processes over 200,000 mobile money transactions per day, applying a combination of rule-based filters and ML anomaly detection models to flag suspicious activity with sub-100ms latency. The pipeline is built on Apache Kafka for event streaming, Apache Spark Structured Streaming for real-time processing, and a custom Python scoring service deployed on AWS ECS. Since deployment, the system has reduced fraudulent transaction losses by 34% year-on-year. has also built and deployed three machine learning models in production environments. At Cellulant, he developed a credit scoring model that analysed alternative data sources — airtime top-up frequency, USSD activity patterns, and mobile money transaction history — to generate credit scores for thin-file borrowers who lack traditional banking history. This model improved loan approval accuracy by 28% compared to the previous manual review process and enabled Cellulant to extend credit to an additional 45,000 borrowers in its first year of deployment. He is equally comfortable with model monitoring, implementing drift detection pipelines using Evidently AI to ensure model quality over time. holds a Master of Science in Computer Science from Makerere University, where his thesis focused on predictive analytics for smallholder farming — specifically, using satellite-derived vegetation indices and historical weather data to predict crop yield outcomes for smallholder maize farmers in Uganda. This work was cited in an FAO working paper on digital agriculture for Sub-Saharan Africa. He is proficient with cloud-based ML platforms, having used AWS SageMaker for model training, deployment, and monitoring, as well as GCP Vertex AI for experiment tracking and batch prediction pipelines. His data visualisation skills include Tableau, Metabase, and Streamlit, which he uses to build self-serve analytics dashboards for business stakeholders. has a strong grounding in data governance, privacy, and compliance. He has implemented data anonymisation and pseudonymisation strategies in compliance with Uganda's Data Protection and Privacy Act and Kenya's Data Protection Act, and he understands the regulatory landscape that data practitioners must navigate across East Africa. In 2024, presented a talk at PyCon Africa titled "Building Production ML Systems on a Startup Budget: Lessons from East Africa," which was attended by over 400 data professionals from across the continent. He is an active member of the Kampala Data Science community and mentors junior data analysts transitioning into engineering roles. is actively seeking data engineering, ML engineering, or AI product roles. He is particularly passionate about applying his skills to fintech, agritech, climate data analytics, and public health data projects where the social impact potential is highest. He is fluent in English, Luganda, and Swahili.