Insights
What we measured, not what we assumed
Notes from building an AI platform over a live ERP: the architecture decisions, the failures that produced them, and the numbers behind both.
Why an SAP AI has to be implemented
A generic assistant pointed at SAP knows how SAP ships. It does not know what your pricing procedure was extended to do in 2014, and no reasoning recovers that.
ReadWhat an AI should say when it fails
Most of the risk in an AI system is not what it does when things work. It is what it reports when a read came back short or the system was unreachable.
The dashboard that froze, and nobody noticed
Eight saved dashboards carried hardcoded date ranges. Nothing errored, no alert fired, and the titles still said last 30 days. Here is why that happens.
Every figure should carry its source
A number without provenance is an assertion. Printing the entity, field, filter and period under every figure is cheap to build and changes what a figure is for.
The SAP report backlog is a symptom, not a workload
Every custom report request is a question somebody could not ask. Clearing the queue faster does not help, because the underlying demand has no bottom.
Why we never let a model write the SAP query
A plausible query against the wrong entity returns a plausible number, and nothing downstream can tell it is wrong. Here is what we do instead, and what it costs.
Browse by topic
Bring the question your reports cannot answer
Thirty minutes against a live SAP system we provide: no access to yours, nothing to set up. If it cannot answer, you find that out in half an hour rather than three months into a pilot.