The teaser on this site is the whole point: a quality of earnings review will not tell you whether the platform is a stack you can sit beside, or a monolith you have to swallow.
Architecture-minded investors look past the product catalogue. They ask whether the firm can overlay intelligence on systems of record it does not own — trust accounting, custody, CRM, document management — through one governed doorway, or whether every new capability requires rewriting the core.
Composable vs disguised monolith
A composable fiduciary stack has a Bridge: read first, write only after a named human approves. Agents are composed from an Action Catalog. Review tiers are versioned. The Record is a per-account hash chain, not a log file that can be edited quietly.
A monolith in disguise looks modern on the diligence slide. The tell is substitution: the vendor wants to become the trustee of record, the custodian, the portfolio system, or the compliance file. r.Team's map is the opposite. The platform overlays. It does not replace.
What to inspect
- Does the architecture name a single auditable doorway to systems of record, or a dozen private integrations?
- Can agents prove themselves against live data before they write — Parallel-Run Independence as a platform control, not a slide?
- Are fiduciary duty, ethics, compliance, cybersecurity, and advisory regulation gates inside the agent loop, or wrappers added afterward?
- Does a public fingerprint leave the boundary, or a client file?
- When the best people leave, does the judgment leave with them as a document they own — or did the institution absorb it?
Those questions live on the Platform, in Vocabulary, and in the NIL Intelligence Likeness paper. Financial due diligence will not ask them. Architecture-minded investors should.