Both dashboards say 1,233. My AI gave itself away with nine decimal places. The βgrey lineβ fooled me too: a 2px blank in 21px of space. Journey 05: teaching a Tableau card how to look right. #Tableau#DataFam#AIAutomation#AgenticBI
Illustrated presenter and synthetic narration. Music: βChill Waveβ by Kevin MacLeod (https://t.co/ar2rnNwHMe), edited, CC BY 4.0: https://t.co/nY8co90zA2
Book-scanner footage: Petr Urbanec, trimmed and sped up, CC BY-SA 3.0: https://t.co/QFRYPy4UqC
Can an AI agent learn a Tableau chart from a workbook?
Self-driving Tableau Desktop Β· Journey 04
7:19 β 11/12 charts rendered.
7:29 β 12/12. Same recipe. I cleared yesterdayβs workbook from the test folder.
Journey 05: every number is right. Why does it look wrong? #Tableau
17 checks passed. Tableau Desktop still couldnβt open the workbook.
One missing connection. After the fix, Ship Mode disappeared. The other sheet stayed.
Journey 03 Β· #Tableau#datafam#aiautomation
Music: Chill Wave β Kevin MacLeod (edited) https://t.co/nY8co90zA2
My Tableau Desktop agent typed into Terminal. Right coordinates. Wrong window.
The fix, three bars, and the check that convinced me: Furniture = 754,748 in both Tableau and the source. #Tableau
Music: Chill Wave β Kevin MacLeod (edited) https://t.co/nY8co90zA2
I gave Tableau Desktop an AI driver. The first surprise? Show Me was a web page. The second? I sent its treemap command myself. Watch the worksheet.
Journey 01 Β· #Tableau#DataFam
Music: Chill Wave β Kevin MacLeod (edited), https://t.co/nY8co90zA2
Behind the glass.
Tableau charts and controls + an externally hosted video backdrop. Would you use motion in a dashboard?
https://t.co/udzPeD6nKQ
#Tableau#DataFam
The build story behind VOTD #2. Two hours on my HTML prototype; agents took longer on the Tableau build.
What would you delegate?
Synthetic voice.
Chill Wave / Kevin MacLeod https://t.co/ar2rnNwHMe CC BY 4.0 https://t.co/nY8co90zA2 Β· edited
This #VizoftheDay by @creativerrr explores NYC Road Safety, analyzing where collisions happen and examining the contributing factors behind them to provide actionable insights for urban traffic safety and operational planning.
Explore the viz on Tableau Public: https://t.co/1XdhZcaWfu
I finished my casual shift at 12:30am and had the interview at 3pm that afternoon. Before it, I asked my agents to publish.
My lesson: let agents build, validate AND publish. I set the direction and review the results.
https://t.co/oJNMkgVPgi
Up at 4am fishing with my son. Six yakka, no kingfish π£
Came home to my Mac mini and my SECOND #VizOfTheDay ππ
NYC Road Safety, built with my Agentic Tableau workflow.
Thank you @tableaupublic, #DataFam and my friends β€οΈ
This #VizoftheDay by @creativerrr explores NYC Road Safety, analyzing where collisions happen and examining the contributing factors behind them to provide actionable insights for urban traffic safety and operational planning.
Explore the viz on Tableau Public: https://t.co/1XdhZcaWfu
Next morning, 4β6am, I built an HTML prototype, then handed the Tableau build to my Agentic BI team.
My hands-on time on the prototype: 2 hours. The agentsβ build and validation time: definitely longer π
Sydney 431 crashes. Cowan KSI 92/150 (61.3%). Counts vs severity.
TfNSW Β· 2021β25; not trip risk.
https://t.co/F7tUjSK0CX
Which suburb should we investigate next?
Drive through Bankstown?
Chapel Rd: 67 of 405 injury crashes (~1 in 6), 2021β25.
Tableau story copy. Which street next?
AI setting/voice. TfNSW data.
https://t.co/F7tUjSK0CX
Music: Chill WaveβKevin MacLeod, https://t.co/ar2rnNwHMe, CC BY 4.0 https://t.co/nY8co90zA2 Β· edited
The top five are only half the story.
Top 5 β Bottom 5 β All 17. Winners or lossesβwhich would you investigate first?
https://t.co/fhFlfMJziU
#Tableau#DataFam