Performance Engineering Python Packages

flameox

Runtime evidence that helps agents trace, profile, and burn down hotspots in application and native code, GPU kernels, and inference stacks.

cpu-perf

A reading path for CPU performance engineering, from one instruction to production inference. Primary sources only, with a runnable benchmark for every section.

mlos-core

MLOS is a project to enable autotuning for systems.

agentloop-profiler

Performance engineering and observability for AI agents.

qaboard

Experiment tracker: organize, visualize, compare and share runs. Removes toil from algorithm/performance R&D and tuning.

minimal-dataset-pytorch

Lightweight experimental PyTorch data-loading library with Parquet-backed datasets, concurrent loading, benchmarking and performance instrumentation.

cuda-engine

Plain English + a PyTorch reference → a verified, benchmarked CUDA kernel you can pip install and torch.compile.

mlos-bench

MLOS is a project to enable autotuning for systems.

mlos-viz

MLOS is a project to enable autotuning for systems.

tuning-metrics

Library to compute auto-tuning and performance metrics.