All posts · page 7

Selected results from the 67-tool prod-code AST suite on apple/swift-argument-parser, as reported: 30,000 lines of Swift, 64 matched throw sites, and remote Apple Silicon test offloading.

Selected results from the 67-tool prod-code AST suite on apple/swift-collections, as reported: 71,000 lines of Swift, 1,279 matched assert and precondition calls, and remote Apple Silicon test offloading.

Selected results from the 67-tool prod-code AST suite on microsoft/TypeScript, as reported: 264,411 declarations across 31,433 files, AST clone clusters, transitive slicing, and compiler in-memory validation.

Historical Zig compiler and standard library analysis with prod-code: source distribution, x86_64 codegen clones, errdefer patterns, and a remote ZLS refactoring example. Zig is now hosted on Codeberg.

A struct with eleven fields shows ten, so any test value built from that answer is missing one, and the only hint is an ellipsis the agent ignored. What it takes to generate a fixture you can trust: read the declaration rather than the summary, and type-check the result before handing it over.

We subjected quickwit-oss/tantivy to all 67 prod-code tools on an idle 32-core cluster node. Reported results: 7,156 module cycle paths, 18-file AST renames, clone detection in JIT expressions, and safe-delete invariants.

We reassessed our September ripgrep report. Its published excerpts do not substantiate the claimed tool coverage or speedups; issue #735 documents a narrower failure in prod-code.

We pointed all 67 tools of prod-code at tokio-rs/tokio on an idle 32-core cluster node. From 6 reported dependency cycle paths and silent Deref evaluation ordering traps to an unhandled sysroot crash on Box-here is what happened.

An agent asked to change one function opens three files and spends most of its context on code that has nothing to do with the task. A slice is the other shape of the same question: the declarations the function depends on, and nothing that merely lives beside them.

A structural rewrite that found nothing, twice, for two unrelated reasons in two different layers. What it takes to tell an empty answer from a broken one, and why the fix was tracing rather than thinking harder.

An agent that knows what it wants but not what it is called has nothing to ask. We gave it a question box, then looked at what the answers tracked in a five-question sample: not the ranking, which was the same everywhere, but the prose your team did or did not write above each function.

A market-data ring in /dev/shm now has identical copies on other hosts, same sequence numbers, same order: about 30 µs across a LAN by multicast, and 51.7 ms p99 across the Pacific with a clock offset good to a few milliseconds.

The agent harnesses we run use the same loop: write a candidate fix into your checkout, run the tests, revert, try the next one. We moved that loop into overlay shadows of the mirrored workspace, mounted at the workspace's own path: candidates run in parallel and the checkout stays untouched.

A 32-core node sitting at 2% load, a one-millisecond network, and half the queries taking 40 ms. Three things we found while measuring instead of guessing, including one optimisation that did not work and one that deadlocked.

Prod keeps agents' terminals on a server and shows them on a Mac and an iPhone. Three emulators disagreed about the same bytes, so we replaced them with one Rust engine that moves its state instead of a picture of it.