AI request cost is visible separately from the path that produced it.
Engineers need actionable request context while the first version must avoid becoming a broad cost product.
Who you’d be doing this for
“Finance asked why tokens jumped last week, and I had to stitch the answer together from three different places.”
Jabari Tekle · Staff Software Engineer
Owns an AI-assisted support workflow with model calls distributed across several services.
What is at stake
Design partners cannot connect rising token use to the request paths responsible for it. You have to weigh a focused investigation view against the pull toward a broader cost-management product.
Why it isn’t already fixed
Every obvious fix costs something else. That’s the part you’d have to decide.
- request-level diagnosis vs. broad cost analysis
- signal availability vs. customer expectations
- fast learning vs. durable product scope
- APM workflow coherence vs. separate analytics surfaces
Why Datadog
Datadog APM already follows distributed request paths that engineers use to diagnose latency and failures.
Written with these in mind
Not your kind of problem? 40 more at Datadog, 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.