Every commit scored for change-risk against this repo's own history, so 'elevated' means elevated here rather than on some global curve.
Needs review
165 commits sit in this repo's top risk tercile, which is 33% of the 500scored. The cut is drawn against this codebase's own history rather than a global curve, so a quiet repo still fills its top band, and here it starts at 8.9 out of 10. What pushes a commit up is size and spread together: a large change confined to one area scores below a smaller one scattered across a dozen files.
Commit categories over time, read off the subject line. Fixes carry the accent because that is the series this chart exists to show.
Consistently other-driven across its history.
Ranked by change-risk, highest first. Priority is a tercile of this repo's own distribution, so a quiet repo still fills its top band.
| # | Commit | Author | When | Lines | Risk | Top driver |
|---|---|---|---|---|---|---|
| 1 | ed690770Move execute-parent kernels into session registry (#8482) | Nicholas Gates | 1mo ago | +529.3K -0 | 100%Elevated | more lines added than baseline |
| 2 | a0a0b207Add layout scan physical plan model (#9142) | Joe Isaacs | 1d ago | +2.6K -0 | 99%Elevated | more lines added than baseline |
| 3 | b51140ad`ListLayout` baseline (#8706) | Matthew Katz | 4w ago | +2.6K -8 | 99%Elevated | more lines added than baseline |
| 4 | 56dffccaNew C++ API (#8741) | myrrc | 4w ago | +2.8K -32 | 99%Elevated | more lines added than baseline |
| 5 | 11f15a90Implement compare in vortex instead of falling back to arrow (#8661) | Robert Kruszewski | 1mo ago | +2.4K -624 | 99%Elevated | more lines added than baseline |
| 6 | 5a096b54Share fixed-width structural array operations (#8864) | Connor Tsuiclaudeassisted | 1w ago | +1.7K -1.8K | 98%Elevated | more lines added than baseline |
| 7 | 3d372165Add a Normalized encoding for tensor columns (#9138) | Connor Tsuiclaudeassisted | 1w ago | +1.9K -1.4K | 98%Elevated | more lines added than baseline |
| 8 | 12fe696eExectuion and optimization tracing harness for testing (#7814) | Adam Gutglickclaudeassisted | 1w ago | +2.9K -66 | 98%Elevated | more lines added than baseline |
| 9 | a6b45278benchmarks: add string compression benchmark (#9060) | Francesco Gargiuloclaudeassisted | 1w ago | +2.1K -76 | 98%Elevated | more lines added than baseline |
| 10 | eafba6c6Add support for huggingface datasets to vortex-python (#8733) | Robert Kruszewski | 2w ago | +2.6K -17 | 98%Elevated | more lines added than baseline |
| 11 | 46b0b2b2`Decimal` Add/Sub kernels (#8724) | Matthew Katz | 3w ago | +1.6K -411 | 98%Elevated | more lines added than baseline |
| 12 | b7ee8736Revert "Lazily untranspose DeltaArray validity and depend on fastlanes bit transpose code" (#8883) | Joe Isaacs | 3w ago | +1.8K -488 | 98%Elevated | more lines added than baseline |
| 13 | c7c28d8cfeat(geo): add native `LineString`, `MultiPoint`, `MultiLineString` types (#8679) | Nemo Yu | 1mo ago | +1.5K -2 | 98%Elevated | more lines added than baseline |
| 14 | 5a764e68assert_arrays_eq and assert_nth_scalar require ExecutionCtx (#8509) | Robert Kruszewski | 1mo ago | +4.0K -2.7K | 98%Elevated | more lines added than baseline |
| 15 | 38251b37Expand the Python expression API (#9375) | Robert Kruszewski | 23h ago | +1.6K -79 | 96%Elevated | more lines added than baseline |
| 16 | 329bba2eSQL style null-on-empty semantics for Sum and Mean (#9113) | Matthew Katz | 6d ago | +1.9K -370 | 96%Elevated | more lines added than baseline |
| 17 | 695e068aAdd Map compute and compression support (#9110) | Adam Gutglick | 1w ago | +1.6K -125 | 96%Elevated | more lines added than baseline |
| 18 | e529ce01chore(deps): lock file maintenance (#8991) | renovate[bot] | 2w ago | +947 -936 | 96%Elevated | more lines added than baseline |
| 19 | c0d1a4e2Add file-level user metadata (#8740) | Martin Prammerclaudeagent | 3w ago | +1.4K -58 | 96%Elevated | more lines added than baseline |
| 20 | 85ed8e41Add `UnionValue` and `UnionScalar` (#8838) | Connor Tsui | 3w ago | +1.4K -57 | 96%Elevated | more lines added than baseline |
| 21 | add69447feat: exploit PiecewiseSequence for faster take (#8802) | Dan King | 3w ago | +1.1K -82 | 96%Elevated | more lines added than baseline |
| 22 | 81c8f136Use PiecewiseSequence indices for FSL take (#8800) | Dan King | 3w ago | +1.2K -277 | 96%Elevated | more lines added than baseline |
| 23 | f742e8a5feat(cuda): export FSST directly to Arrow varbin (#8787) | Alexander Droste | 4w ago | +1.1K -176 | 96%Elevated | more lines added than baseline |
| 24 | 198cbd5cNew C++ API tests (#8751) | myrrc | 4w ago | +1.6K -85 | 96%Elevated | more lines added than baseline |
| 25 | e519a02cReorganize the compressor (#8745) | Connor Tsui | 4w ago | +1.6K -1.6K | 96%Elevated | more lines added than baseline |
| 26 | 42bea5deAdd golden-corpus determinism tests for the default compressor (#8697) | Connor Tsuiclaudeassisted | 4w ago | +1.2K -0 | 96%Elevated | more lines added than baseline |
| 27 | 0d503423Replace ToCanonical trait usages with execute (#8609) | Robert Kruszewski | 1mo ago | +2.2K -1.5K | 96%Elevated | more lines added than baseline |
| 28 | 97850e9eAdd a dormant v4 RDS Postgres dual-write to the CI benchmark emitters (#8513) | Connor Tsuiclaudeagent | 1mo ago | +1.3K -27 | 96%Elevated | more lines added than baseline |
| 29 | 4187c1a9Use aggregate descriptors for zoned stats (#7938) | Nicholas Gates | 1mo ago | +1.8K -645 | 96%Elevated | more lines added than baseline |
| 30 | 7e06a99bVarBinViewBuilder: support decoding Dict and Sparse arrays directly into the builder (#9236) | Robert Kruszewskiclaudeassisted | 6d ago | +924 -24 | 92%Elevated | more lines added than baseline |
| 31 | 74a2b861refactor(vortex-geo): generalize scalar function execution (#9215) | Nemo Yu | 6d ago | +777 -521 | 92%Elevated | more lines added than baseline |
| 32 | 0f0390a6Improve take on Chunked(FSL) and FSL(Chunked) array alongside implementation of take using PiecewiseSequence (#8881) | Robert Kruszewskiclaudeassisted | 6d ago | +1.1K -46 | 92%Elevated | more lines added than baseline |
| 33 | 6a8d2484refactor: preserve bound expression identity through scans (#9192) | Joe Isaacs | 6d ago | +1.7K -767 | 92%Elevated | more lines added than baseline |
| 34 | a1057db5FFI: get scalars from arrays (#8896) | myrrc | 1w ago | +891 -450 | 92%Elevated | more lines added than baseline |
| 35 | 93058ba6Lazily untranspose DeltaArray validity take #2 (#9226) | Robert Kruszewskiclaudeassisted | 1w ago | +820 -2.0K | 92%Elevated | more lines added than baseline |
| 36 | c8dc9039Add Arrow Map roundtrip support (#9111) | Adam Gutglick | 1w ago | +937 -45 | 92%Elevated | more lines added than baseline |
| 37 | bc9aa104Add Map logical type and scalar support (#9107) | Adam Gutglick | 1w ago | +1.8K -25 | 92%Elevated | more lines added than baseline |
| 38 | fd43ebd4feat: support bound expressions in partitioning (#9157) | Joe Isaacs | 1w ago | +798 -38 | 92%Elevated | more lines added than baseline |
| 39 | 452ef6a7feat(arrow): export StringArray and BinaryArray directly (#8902) | ClSlaid | 2w ago | +1.2K -106 | 92%Elevated | more lines added than baseline |
| 40 | 1e6102bebreak: new layout tree vtable (#8910) | Joe Isaacs | 2w ago | +1.4K -1.8K | 92%Elevated | more lines added than baseline |
| 41 | c5e075abAdd PiecewiseSequence index array (#8799) | Dan King | 3w ago | +630 -0 | 92%Elevated | more lines added than baseline |
| 42 | 54b045b3feat(jni): bridge caller-provided Java I/O into VortexReadAt/VortexWrite (#8748) | Robert Kruszewski | 4w ago | +1.3K -0 | 92%Elevated | more lines added than baseline |
| 43 | 7f8d92f0VortexArrowColumnVector supports conversion from all arrow types that Spark can support (#8771) | Robert Kruszewski | 4w ago | +947 -198 | 92%Elevated | more lines added than baseline |
| 44 | 8a1f91c0feat(vortex-geo): geometry `Bbox` zone-map statistic + distance-filter pruning (#8646) | Nemo Yu | 4w ago | +1.5K -21 | 92%Elevated | more lines added than baseline |
| 45 | 002e40afST_Intersects pushdown: vortex.geo.intersects kernel, DuckDB lowering, spatial function overrides (#8704) | Nemo Yu | 4w ago | +1.1K -259 | 92%Elevated | more lines added than baseline |
| 46 | a530b3b7Add `UnionVariants` type and embed it in `DType::Union` (#8703) | Connor Tsui | 4w ago | +1.3K -38 | 92%Elevated | more lines added than baseline |
| 47 | b1f2d4f0Cast pushdown for duckdb (#8620) | myrrc | 1mo ago | +1.1K -271 | 92%Elevated | more lines added than baseline |
| 48 | 890704f4feat: add CUDA cuDF convenience API (#8624) | Alexander Droste | 1mo ago | +915 -34 | 92%Elevated | more lines added than baseline |
| 49 | 0a45777eAdd swimlane view to vortex-web that visualizes physical layout of columns in files (#8634) | Robert Kruszewski | 1mo ago | +683 -238 | 92%Elevated | more lines added than baseline |
| 50 | 3e7098cffeat(vortex-bench): infra of SpatialBench on DuckDB, plain binary (#8598) | Nemo Yu | 1mo ago | +1.3K -299 | 92%Elevated | more lines added than baseline |
Two views of the same model: where the cuts fall, and what commit shape lands you above them.
Every scored commit, binned on the raw 0 to 10 score rather than the percentile. Percentile ranks are uniform by construction, so that axis has no shape to draw. The dashed lines are the tercile cuts behind each row's priority pill.
The 200 most recent commits, on their own recency sample rather than the feed above: that defaults to risk-sorted, so reusing it would plot only the top tercile and call it the spread. Big and scattered is what the model penalises. Click a dot to open it.
vortex-data/vortex has 500 commits in its history from 33 contributors, the first of them Jun 18, 2026. In the last 90 days 2,335 files were touched, 5,936 times in total, most often tests.rs. Every commit is scored for change risk from its size, spread and the history of the files it touches.
