Seek PhoenixAthena
A multi-agent AI pipeline for SAP development — drafts specs, generates ABAP code, and debugs, with a review step before anything ships.
Writing a functional spec, then a technical spec, then the ABAP code to match, then debugging it — that's slow, repetitive work spread across several roles and handoffs. Athena is an agentic pipeline where AI agents handle each stage: drafting the FDD and TDD from a requirement, generating the corresponding code, and working through debugging — with a review step built in before anything is signed off.
Requirement comes in
A business or technical requirement is handed to the pipeline as a starting point.
Agents draft the specs
One agent produces the functional design document, another the technical design document, grounded in the requirement.
Code is generated
ABAP code is generated to match the technical spec, ready for the review stage.
Review and debug loop
A review agent checks the output and cycles through debugging before anything is marked ready.
Requirement in → spec and code drafted by agents → reviewed → build out.
Stack
This describes the public-facing pipeline only. The fully local version isn't shown here.