Postmortem Pile
Cross-references five years of incident reports to find the failure you're about to repeat. Built for Receipts โ build something that makes a messy pile of documents answerable. Similarity search across a real corpus of 340 reports. The interface is deliberately narrow: one path, done properly, rather than a broad surface that half-works at the deadline.
Gallery
Screens
Postmortem Pile โ main view
Postmortem Pile โ result state
On the clock
What shipped, what didn't
Completed in time
Similarity search across a real corpus of 340 reports.
Unfinished / faked
Ranking is embedding-only, no recency weighting.
Obstacles
Lost four minutes to an auth redirect loop that turned out to be a trailing slash.
Lessons learned
Boring, well-understood tools beat interesting ones when there's no time to debug.
Process
Prompting approach
Started from a single system prompt describing the user and their constraint, then iterated with the assistant on the Postmortem Pile data model before writing any interface code. Full prompt log is in the repository under /prompts.
Fairness
Disclosures
Declared by the builder at submission time, visible to judges before scoring.
Judging
Public feedback
Judges' private notes stay private โ this is what they chose to say publicly.
Elena Voss
VP of Engineering, Harbor Compute
Nice work under the clock. The unfinished list is longer than the finished one, but what shipped works and you were straight about it.
Claudia Reyes
Staff Engineer, Vantage Analytics
Nice work under the clock. The unfinished list is longer than the finished one, but what shipped works and you were straight about it.
Hana Kobayashi
Principal Research Engineer, Independent
Solid build with a clear user in mind. The interface is doing less than it could โ a second pass on hierarchy would lift this considerably.
Discussion
0 comments
No comments yet.