A private, invitation-only streaming room where Fundacion Rimas shares music by its youth participants with funders and allies, without ever handing anyone a file.
Fundacion Rimas runs La Academia, a music, arts and sports program for Puerto Rico's most vulnerable youth; Mutiny Labs already built its participation and impact platform. The program's students write and record real songs, and the foundation needs funders, partners and families to hear them. But sending audio files of a minor's creative work means losing control of it forever.
Every mainstream option was wrong. Public platforms expose children's identities and invite reuse. Email attachments are unrecoverable the moment they leave the outbox. The foundation needed listeners to feel like guests at a listening session, not recipients of files, with access granted per person, per song, and withdrawn just as easily.
We treated the player like a vault with a beautiful lobby. Audio is transcoded and encrypted in the administrator's browser, so the original recording never leaves their computer unprotected. Every playlist, decryption key and six-second audio segment is served through an authenticated API that re-checks identity and per-song permission on every request. The interface itself implements the foundation's 2026 brand system, from typography to the red play triangle.
Audio lives as AES-128 encrypted six-second segments in private storage. The browser only ever receives a rewritten playlist, a session-checked key and non-cacheable encrypted segments. No audio file ever reaches a listener's device.
Admins invite listeners by email and choose exactly which songs each one can hear. Revoking access takes effect on the next segment request, cutting off a stream in about six seconds, even mid-song.
Every login, visit and play is recorded with IP address and device, alongside per-listener listening minutes. The foundation can always answer who heard what, when, and for how long.
Generative AI transcribes each track into time-synced lyrics that highlight karaoke-style during playback, so listeners can follow every word the young artists wrote.
When a track has no artwork, generative AI reads its lyrics and paints an abstract album cover in the brand's spirit. Prompts exclude faces, text and any depiction of children by design.
under the hood: Server-rendered React and TypeScript, AES-128 encrypted HLS streaming, in-browser WebAssembly audio encryption, managed Postgres and private object storage, generative AI for lyric transcription and cover art, serverless API layer on a global edge network
We reply personally. A senior partner, not a sales pod. We'll tell you straight whether we can help.
