Skip to content
RYAN BRENTS / SERVICES / DOCUMENTATION
●LIVE · ATL

Documentation

The vendor documents the product. What usually goes undocumented is your copy of it: the custom tables, the DML someone wrote in 2011, the integration only one person understands. I rebuild the map — the schema, the screens, the programs, the data flows — until the system is legible again, to your team and to your AI tools.

THE WORK — /DOCUMENTATION
01System reconstructionReverse-engineer a schema, customizations and all, into a searchable data dictionary — every table, column, key, relationship, and valid-value list, with field-level lineage. Your system as it actually runs, not as it shipped.
02Knowledge graphs & schema mapsTurn thousands of tables into something you can actually navigate: a force-directed graph, path-finding between objects, and module grouping — so 'how does this connect to that' takes a click, not a career.
03Process & integration docsDocument EMF interfaces, alerts, and integrations end-to-end from their real configuration — the node graph, the wiring, the failure modes — instead of from whoever still remembers building them.
04Web services & API surfaceMap every integration entry point: SOAP/Connect web services with typed parameters, field mappings, and the tables behind each one. The surface your integrators keep asking about — 233 of them on rossdata.dev alone — in one place.
05Machine-readable outputNot a 400-page PDF nobody opens. A REST/JSON API and an MCP server, so your systems, your dashboards, and your AI tools can all query the documentation directly.
THE PROOF — IN PUBLIC

I documented all of standard Ross 8.0. Then I put it online.

1,840 tables, 36,230 columns, 233 web services — a public data dictionary, knowledge graph, REST API, and MCP server for a system almost nobody has mapped in full.

SEE THE WORK →
Documentation — signal-routing art
PROOF IN PUBLIC — LIVErossdata.dev

The Ross ERP 8.0 schema, made legible.

The entire standard Ross ERP 8.0 schema, documented and put online: a searchable data dictionary, an interactive knowledge graph, 233 typed SOAP/Connect web services, a public REST API, and an MCP server your AI tools can query directly. Proof I know this system down to the column.

OPEN ROSSDATA.DEV →
233
WEB SERVICES
1,840
TABLES
36,230
COLUMNS
2,327
SCREENS
THE WRITING — RELATED
2026-07How to think about modernizing Ross ERP in 2026→2026-07RAG for manufacturing documentation→
SEE ALSO

sitting on a system only one person understands — one email starts it

NO CALENDAR LINKS · NO FUNNEL · JUST MAIL

rossdata.dev covers the standard Ross schema. Point the same toolchain at your own instance and you get a private version — yours, behind your login.