As part of P&G's innovation push in smart grooming devices, the Grooming R&D team needed a scalable system to process and analyze time-series data from sensor-enabled razors. With each new consumer study introducing changing device firmware and evolving data structures, the solution had to be flexible, maintainable, and fast.

I engineered robust data pipelines within Databricks, using PySpark and SQL, to handle the full lifecycle of study data—from raw ingestion to dashboard-ready insights.

• Built modular pipelines following the medallion architecture (Bronze → Silver → Gold), transforming raw sensor data into analytics-ready tables.
• Integrated the pipelines with Azure Cloud Storage and structured them as Delta Live Tables for scalability and real-time refresh via scheduled workflows.
• Designed and launched a new Health Dashboard in Power BI, enabling study owners to monitor key metrics with clean, interactive visuals.
• Improved dashboard UX by merging redundant study pages, simplifying navigation, and optimizing underlying data queries.

• Delivered 10+ complete study cycles, each with customized data handling and visualization.
• Proactively refactored existing pipelines, improving efficiency and adaptability to firmware changes.
• Enhanced team agility with clear sprint-based task management in Jira and visual planning in Mural.
• Improved data accessibility and stakeholder satisfaction by streamlining complex dashboards into a cohesive user experience.

This project shows my ability to turn complex, messy data into actionable insights, support real-world product innovation, and work cross-functionally in a fast-paced R&D environment. It's a perfect example of how thoughtful data engineering can directly power smarter consumer products.