fix: route resume grading through AI Gateway to stop false "future dates" flag (SIL-3430)#18
Open
LuchoQQ wants to merge 1 commit into
Conversation
…tes' flag (SIL-3430) The google() direct serving of gemini-2.5-flash intermittently flags past 'Present' start dates as being in the future, dragging excellent resumes down to grade C. The same model via the Vercel AI Gateway does not reproduce this (0 hallucinations across an isolated same-code A/B). Route grading through the gateway. Requires a paid AI_GATEWAY_API_KEY in prod.
|
@LuchoQQ is attempting to deploy a commit to the conanbatt's projects Team on Vercel. A member of the Team first needs to authorize it. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
gemini-2.5-flash flags past dates ("Oct 2024 - Present", "09.2024 - Presente") as being "in the future". It correlates 1:1 with grade C and drags excellent resumes down from S to C with a false accusation.
It depends on the serving backend, not the prompt. The same model served via
google()direct (Google AI Studio, what prod uses) hallucinates; served via the Vercel AI Gateway, it does not.