feat(eval-view): optimize dashboard load time with GCS summary manifest - #1179
Merged
Conversation
… suites.gen.json - Extract compact suite summaries containing guided/unguided stats, rates, guides breakdown, and chartData via extractSuiteSummary - Refactor landing.js and guide.js to load remote suites in a single HTTP request - Update upload_suite.ts to automatically reconcile and upload suites.gen.json to GCS on upload with prefix pagination - Update server.js and parity.test.ts for SuiteSummary schema
…trim unused summary properties
paulirish
approved these changes
Aug 5, 2026
paulirish
left a comment
paulirish
left a comment
Member
There was a problem hiding this comment.
yeah eval-view is the part of our codebase that's all untyped and it scares me. but.. thats more my fault and problem. so.. in the future ill be happy to have proper types across these objects. :)
but overall. .. agree on upgrading suites.gen.json to have all this data and solve the problem. the reconcile looks god
yup love it. thanks
paulirish
approved these changes
Aug 5, 2026
# Conflicts: # eval-view/landing.js
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Alternative to #902. Instead of speeding up by paginating results, I figured it'd be faster to store a
suites.gen.jsonfile in GCS which contains all the metadata needed to populate the landing page of the dashboard. This file gets updated whenevergd uploadis called.@paulirish let me know what you think of this approach!
Summary
Replaces individual per-suite
evals.jsonfetches during remote results loading with a single compact summary manifest (suites.gen.json) stored in GCS.Impact
Key Changes
eval-view/summary-extractor.js: Extracts compact suite summaries (guided/unguided stats, rates, per-guide breakdowns, and hoverchartData) and metadata necessary for filtering (ID, agent, model, source, etc.).eval-view/landing.js&guide.js: RefactoredloadRemoteTests()to fetchsuites.gen.jsonin a single request and consume pre-computed summary fields directly.harness/upload_suite.ts: Automatically updatesgs://guidance-evals/suites.gen.jsonin GCS ongd uploadwith prefix pagination and parallel backfilling.eval-view/server.js&parity.test.ts: Updated/api/suitesendpoint and parity tests for the summary manifest schema.Verification
pnpm preflight(build, typecheck, lint, and tests all passed).suites.gen.json(156 suite summaries) to GCS.