docs: performance + correctness-evals roadmap - #7
Draft
riwsky wants to merge 1 commit into
Draft
Conversation
Deep pass on the remaining hot paths (AX-walk XPC storm, selector projection, timing constants, pasteboard typing) plus an eval plan in the LLM-evals sense: perf instrumentation, deterministic correctness suite against the playground apps, agent-eval runner for evals.json, and token-efficiency metrics. Includes research verdicts on the AXP multipleAttribute bulk request (the big lever) and cut alternatives. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01UskKrPmBjwvuKUvB1ioN3U
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds two docs capturing a deep planning pass on performance and evaluation — docs only, no code changes.
What's in here
docs/perf-and-evals-roadmap.md— findings + prioritized roadmap:exists/tap/text/wait) pays for the full tree with full attribute projection.multipleAttributebulk request (requestType 5) that iosef can build directly and send over its existingsendAccessibilityRequestpath — ~9 XPC/element → 1, no AXPTranslator-singleton thread-safety concerns.--depth).IOSEF_PERF=1instrumentation (XPC calls per op as a regression metric), a deterministic correctness suite (scripts/correctness.py) with per-machine baselines, a runner for the existingevals.jsonrubrics, and token-efficiency metrics in benchmark.py. All tracking is local JSONL/markdown.docs/perf-and-evals-design.md— per-item designs (function signatures, spike steps, check-by-check details), research verdicts on cut alternatives (xctree's public-AX approach, WDA/XCTest snapshots, translator-concurrent walks), and sources.Suggested first step
Land roadmap items 1–2 (perf instrumentation + correctness suite) before any optimization, capture a baseline with
correctness.py --update-baseline, then gate every perf change on it.🤖 Generated with Claude Code
https://claude.ai/code/session_01UskKrPmBjwvuKUvB1ioN3U
Generated by Claude Code