SYSTEM Cited by 1 source

Numba

Numba is an open-source JIT compiler for Python that translates a subset of Python + NumPy code into optimised machine code via LLVM. Canonically used to accelerate numerical / array-oriented hot paths in data-science code without rewriting in C / Cython.

Stub page — minimum viable framing.

Seen in

• sources/2025-06-29-zalando-building-a-dynamic-inventory-optimisation-system-a-deep-dive — canonical first wiki disclosure as a named speed lever inside Zalando's ZEOS inventory-optimisation pipelines. Verbatim Zalando claim:

"For data and vector transformations, we rely on PySpark, Numba and Joblib multi-core parallelisation. Whenever possible, we vectorise operations and rely on Numba, which can often offer speedup by a factor of 2 or 3 compared to NumPy."

Sits inside the vertical-scaling tier of the two-tier feature-engineering split (concepts/data-preprocessing-vs-data-transformation-split) — used in the SageMaker Processing Job where libraries lack native distribution and bigger instances + JIT acceleration are the lever.

Related

• systems/mlforecast-nixtla — uses Numba under the hood for time-series feature generation.
• systems/sagemaker-processing-job — typical Zalando compute tier.
• systems/zeos-demand-forecaster — Zalando consumer.
• companies/zalando