Breakwater

Codeflash / rendercv#17

⚡️ Speed up function validate_typst_dimension by 59%

IanHollow/rendercv#17 · merged 2026-08-10 · base ddd75cabd6 → head 7f7abe4ebf · ran on Sandboxes

PASS

No behaviour divergence found by any probe.

† Merged after Nemotron 3 Ultra was released (2026-06-04): the model cannot have seen this PR or its fixes.

Performance · validate_typst_dimension

No realistic input family measured slower.

head slowerhead faster0.59×0.71×0.84×1×input size (characters)F2F2: 0.56× at ~8 charsF2: 0.57× at ~8 charsF2: 0.57× at ~9 charsF2: 0.57× at ~9 charsF2: 0.57× at ~9 charsF2: 0.57× at ~9 chars
  • F2
Head/base cachegrind instructions per call, by input size (log-log). Each line is one input family Nemotron proposed; points above 1× are where the "faster" code does more work.
FamilyInputsRatio range Slope base → headVerdict
F2Repeated calls with typical valid Typst dimensions (e.g. "10cm", "12.5pt", "3em")claimed speedup · plan0.56× – 0.57×0.00 → 0.00faster

Receipt

TargetBackendImageRun AuditAdjudication
rendercv.themes.options:validate_typst_dimensionToken Factory Sandboxesaa55391a-3699-4da1-90cb-9549f152a03dbt-rendercv#17-945eef$0.032$0.000

Nemotron cost on Nebius Token Factory. Backend: where the untrusted code ran. Token Factory Sandboxes (a microVM branched from a kept image, cited by its UUID). Every image is content-addressed (repo at each commit + harness), so every card re-runs from the image and seed 0. Raw data: audit.