Andrea Ranzato

• Partnered with pet nutrition expert Alice Chierichetti to launch nutrivetpet.com, reaching 6,000+ users, 1,050+ diet sales, and a 14% conversion rate in 250 days.
• Architected the core nutrition algorithm following veterinary specifications, processing pet biometrics for precise dietary requirements and portion calculations.
• Managed the technical partnership with an external developer, defining system architecture and requirements for a scalable web application in TypeScript, Next.js, TailwindCSS, shadcn, and Prisma.
• Automated administrative tasks with internal R packages: traction reporting and monthly revenue reports via Stripe API integration.
• Developed the open-source R package postmarkr for marketing email automation, and designed the pricing algorithm for the partner sponsorship model.

• Architected high-performance data infrastructure for a Eurasian central bank, leveraging Apache Arrow to process and analyse massive financial transaction datasets and reducing query times by orders of magnitude.
• Contributed to automated data pipelines processing 30+ datasets to generate customised statistical reports for 200+ Swiss municipalities, using DuckDB for in-memory analytics and SQL Server integration.
• Supported climate and energy data workflows handling millions of renewable energy datapoints, implementing an Arrow-based streaming architecture for API integration and dashboard updates.
• Assisted the "R for Databases" workshop at posit::conf(2024) Seattle, creating exercises and supporting 20+ international data professionals learning DuckDB, dbplyr, and DBI best practices.

• Maintained robust Python-based data engineering pipelines for HR analytics for European intergovernmental organisations, automating statistical survey processes and enabling data-driven workforce insights.
• Contributed to open data initiatives for Regione Lombardia.

• Sustained enterprise-wide BI reporting infrastructure using Qlik Sense for a public utility company, establishing automated pipelines and interactive dashboards that enhanced operational decision-making.

• Optimised the automated monitoring system for fraud detection KPIs, streamlining performance analytics across fraud prevention teams and reducing manual reporting.
• Maintained an R-based data integration pipeline for global chargeback analysis, consolidating data from multiple financial acquirers to enable fraud pattern detection.

Acquired theoretical and computational skills (R, Python, SQL) to solve data analytics problems — regression, classification, inference. Courses attended: statistical inference, Bayesian statistics, business analytics and data mining, data management and warehousing, statistical learning, computational statistics, patents and property rights for big data, and dynamic economic analysis. Thesis: "Yield Curve Modelling with Principal Component Analysis for Market Risk Assessment".

Intensive summer programme on advanced statistical methods for risk assessment and management, covering modern techniques in risk modelling, extreme value theory, and quantitative risk management frameworks used in financial institutions.

Built a foundation in economic theory and quantitative methods, with a focus on econometric analysis and statistical applications in economic research — the groundwork for my transition into data analytics. Elective courses: applied statistics, econometrics, mathematics for economics and finance, international economics, monetary economics, and industrial organization.