Faster anomaly flags compete with the need to keep engineers responsible for data holds.
The pilot must surface suspect ads data early without turning an uncertain flag into an automatic downstream decision.
Who you’d be doing this for
“I finished the Quest, then my friend got credit and I’m still waiting for it to show up.”
Ha Rattanakorn · Community Member
Their experience of relevant Quests and ads depends indirectly on reliable delivery and measurement data.
What is at stake
A quality alert is reaching engineers after a dashboard or ML input has already used the affected run. You must weigh faster evidence gathering against keeping every data hold under engineer review.
Why it isn’t already fixed
Every obvious fix costs something else. That’s the part you’d have to decide.
- faster anomaly review vs engineer-controlled data holds
- training freshness vs input quality confidence
- richer evidence vs alert fatigue
- reusable automation vs pipeline-specific thresholds
Why Discord
Ads delivery, attribution, and ML inputs depend on canonical datasets that must remain trustworthy across measurement surfaces.
Written with these in mind
Not your kind of problem? 16 more at Discord, or browse every organization.
This is the setup. The work is inside.
Running it puts you in the room: the full situation and its constraints, stakeholders who push back in their own words, and the decisions that are yours to make. What you produce becomes a Day One Plan — work you can show someone instead of describing.