- Machine Learning
- NLP
- Python
- Azure
- Forecasting
- SQL
- Communication
- Marketing
- Spark
- Ci/Cd
- Pytorch
- Power Bi
- Model Monitoring
- Feature Engineering
- Architected personalisation engine serving 200M+ monthly users, improving recommendation relevance by 35%.
- Improved model observability stack, reducing drift alert noise by 70% and accelerating retraining cycles.
- Partnered with engineering to migrate core prediction APIs from Python Flask to Go, decreasing latency by 50%.
- Led cross-functional workshops on experimentation best practices, increasing test coverage by 40%.
- Mentored two junior data scientists through productionisation of their first customer-facing models.
- Developed automated monthly performance dashboards for retail clients, reducing reporting cycle time by 5 days.
- Created customer segmentation models that informed targeted campaigns, lifting client revenue by 15% on average.
- Streamlined ETL pipelines using Apache Airflow, decreasing data latency from 48 hours to 6 hours.
- Conducted workshops on Tableau best practices, upskilling client marketing teams to self-serve analytics.
- Collaborated with data engineering to establish governance around customer data across multiple sources.
- Designed fraud detection models that reduced false positives by 25% while maintaining 98% detection rates.
- Migrated legacy scoring pipelines to AWS SageMaker, cutting deployment cycle time from weeks to hours.
- Presented findings to executive audiences, translating complex model behaviours into clear business insights.
- Improved test coverage of ML codebases by instituting strict pre-release validation checklists across teams.
- Coached associates on Pythonic data manipulation practices, improving overall team coding standards.
- AWS Certified Machine Learning - Specialty
- Microsoft Certified: Azure Data Scientist Associate