In my current role at Mars, I lead the development of a crucial high-performance analytics web app built with Azure, FastAPI and React (Next.js) for dynamic data visualization (d3, recharts).

I also lead the ingestion and transformation of diverse e-commerce data sources using technologies like Databricks, dbt, and Snowflake. My work is pivotal in driving data-driven decision-making and enhancing user experience across our digital platforms.

Contributed to the development and deployment of a modern machine learning platform for a leading UK commercial bank (Google Cloud, Python).

Spearheaded the development of a custom wrapper for ML models that automatically deploys them using Docker. The solution is being used for all production ML deployments at the bank today.

Designed the solution architecture and data model of a cloud-based OLAP system processing billions in annual debt.

Drove discussions with senior stakeholders to understand complex business processes and codify them into data processes, architectures and systems.

Conducted extensive stress testing by developing pilot-scale ETL pipelines using Jupyter notebooks.

Managed a team of analysts to design, develop and architect cloud-native big data web assets (React, Next, Express) for the GCP.AI practice using agile methodologies.

Led the full-stack development (NextJS, FastAPI) of a groundbreaking Generative AI Toolkit to enhance BigQuery's capabilities, along with a customer scenario simulator in the telecommunications industry.

Developed an award-winning GCP-hosted web app (React, Express, BigQuery) which played a crucial role in a £300k client bid, encouraging responsible gaming.

Developed OLAP ETL pipelines (PySpark, SQL) and analytical reports for a data analytics platform for a healthcare client, expected to generate millions in yearly savings by optimising treatment delivery.

Designed and implemented reporting software to address data quality issues and provide crucial insights for day-to-day management operations for thousands of daily theatre sessions.

Created and deployed a highly effective data visualization software module across all UK branches, improving data analysis and decision-making capabilities.