Investigate yield variation, compare process conditions and connect test measurements to wafer position. l supplies the column operations and numerical expressions for moving from a lot-level summary to the devices behind it.
01 / The workload
A shift can appear across lots, process steps, test bins or regions of a wafer. Understanding it means retaining both the measurement and its manufacturing context. The useful question changes as the investigation moves from a broad trend to one device population.
Group device outcomes by lot, wafer and test classification. Calculate counts and proportions, compare populations and extract the measurements behind a change.
Use die coordinates and recorded measurements to compare regions. Prepare the coordinate-level aggregates needed by your wafer-map visualization and engineering analysis.
Analyze datasets that associate test outcomes with process steps and equipment conditions. Compare measurements across tools, recipes and operating periods.
Calculate statistics across numerical test columns and study relationships between measurements. Isolate outliers or populations for further review in your engineering tools.
02 / From data to analysis
Use the same identifiers and coordinate fields from the manufacturing record through to the focused engineering extract.
Start with flat measurement tables and associated lot, wafer, die, recipe and equipment identifiers.
Filter test conditions, group outcomes and calculate distributions or numerical relationships across the selected devices.
Return comparison tables, coordinate-level aggregates and individual device rows to your quality and engineering applications.
/ Count devices by wafer l>t:([]wafer:1 1 1 2 2 2;ok:110101b) l>select sum ok by wafer from t wafer| ok -----|--- 1 | 2 2 | 2 l>exec avg ok from t 0.6666667 l>exec wafer from t where not ok 1 2 l>
03 / Why l
You write q/k. The runtime considers the physical representation of the data, choosing encoding-aware operations where supported and materializing values when needed.
04 / Your first evaluation
Choose several lots with a known test shift, consistent die coordinates and measurement tables that include both passing and failing populations.
Available as a research preview for local evaluation.
How physical layouts become part of execution.
Read the paper Integration guideCall l from Python, pass typed values and bring results back to your application.
Read the guide Integration guideRead, write and stream Arrow IPC files with the native interop library.
Read the guideYour data. Your next question.