Muhammed Marong

Built and deployed Sisi, a production multilingual AI voice assistant serving 3,000+ smallholder farmers in five languages. Developed ML demand forecasting models achieving 94% R² accuracy and designed the full analytics stack.

Designed and ran A/B experiments across millions of payment transactions to improve conversion rates and reduce funnel drop-off. Built and owned Looker and Tableau dashboards monitoring payment KPIs, retry logic, and failure states.

Designed and delivered a national real-time data portal and early warning system used by police, healthcare, environment, and law enforcement departments. Gathered requirements from multiple government stakeholders and translated them into product specifications.

Oversaw 1,000+ field enumerators covering 200,000–300,000 people during the National Population Census. Monitored real-time data sync, ran consistency checks, identified anomalies, and coordinated corrections.

Analysed in-app behaviour and support interactions for 10M+ users across 12 Asian markets using SQL and Python. Designed and maintained Tableau dashboards tracking CSAT, churn signals, resolution time, and ticket volume.

Built statistical models for disease surveillance and health outcome forecasting on large epidemiological datasets using Python and R. Contributed to national early warning system dashboards across food security, public health, and human security.

Pursued a Master of Science in Data Science with a thesis on ML-driven rice yield forecasting, achieving 94% R² accuracy. Published research in IJACSA 2024.

Grade: First Class Honours

Earned a Bachelor of Science in Computer Science with First Class Honours. Published a peer-reviewed paper on sentiment analysis in e-commerce using NLP in 2020.