Business Intelligence v1: data lake

TrueStory's first attempt at a company-wide data lake, and my role was foundational: research, not execution against an existing plan, because there wasn't one yet.

I did the initial investigation into what building a data lake for all of TrueStory's data would actually take, documented findings and technology-selection recommendations for the team to act on, and then went further than a slide deck, building proof-of-concept solutions to validate the approach before anyone committed real engineering time to it.

The stack under consideration was AWS-native:

• Athena for querying
• Glue and Glue DataBrew for transformation
• S3 as the lake itself
• Lake Formation for access governance
• EventBridge for event routing
• Secrets Manager for credentials

Alongside the technology research, I worked directly on data denormalization, transformation, and enrichment, the unglamorous work of actually shaping raw data into something a data lake could usefully hold. It's the project in my TrueStory stretch that looked least like typical backend work and most like the AI/data-engineering direction my career eventually turned toward.