I'm a BI developer in Ottawa. My background is in pharma (PharmD, 10+ years across production, quality and regulatory affairs), so I usually know what the data means before I chart it.
These days: Power BI (DAX, Power Query), SQL, a lot of Python.
- Dashboards that answer one question well, instead of showing every number I have.
- Data models I'd be happy to hand to someone else: documented code, clean schemas, Power BI projects in PBIP/TMDL so they diff cleanly in Git.
- Honest write-ups: when a model loses to a naive baseline, or the data has a gap, it goes in the README.
- Data science tinkering on the side: whatever dataset catches my interest, from retro game markets to sound engineering.