Overview
Waveflow is a Python-native framework for algorithm, hardware, and software co-design. Instead of fragmenting a hardware system across algorithm notebooks, architecture spreadsheets, simulation harnesses, HDL, software bindings, and build scripts, Waveflow describes it once as structured, executable Python — so simulation, generated implementation, software, and tooling all stay aligned around a single model.
This section covers the why, the what, and the how — plus the concrete system that motivates Waveflow and an honest status:
- Motivation — the problem, Waveflow’s single-source approach, and who it’s for.
- The Python model — what a Waveflow component is: schemas, interfaces, parameters, and a compute hook.
- The Waveflow flow — the two-loop design methodology, end to end.
- The harness for AI — why Waveflow is the substrate that makes AI effective for hardware design.
- SALSA — the reconfigurable wireless system Waveflow was built for.
- Project status — what works today, what’s next, and the first integrated milestone.
New to the project? The fastest way in is the basic vectorization example — one multiply-accumulate, bit-exact from Python to Vitis.