Dev Blog

Technical writing from the team building ONCE.

The ONCE Fraud API: A One-Way Mirror Over the Risk Engine

How we turned an internal fraud-scoring engine into a public API. The design is mostly about what we refuse to return: no vendors, no IPs, no PII, no internal flag ids.

securityengineeringapi

How We Built OMG: AI Music Generation to DSP Distribution in One Flow

A technical deep-dive into ONCE Music Generation. Genre-aware lyrics prompting, the credit surcharge system, AI detection, and the full audit trail from prompt to published release.

aiengineeringmusic-gen

ONCE Fraud Detection: Automated Account Risk Scoring

How we built ONCE Fraud Detection, a system that computes daily account risk scores from audio fingerprints, network signals, streaming patterns, and AI conversation analysis.

securityengineeringinfrastructure

Meet Note

The universal assistant that lives alongside your workflow — in the app, on Discord, and wherever you need help.

aiengineeringagents

The Publishing Agent

An AI agent that auto-drafts publishing metadata using citation-backed web research, strict evidence gating, and deterministic share math.

aiengineeringagents

The Edit Agent

A small agent that lets users edit release metadata by describing what they want changed, no forms required.

aiengineeringagents

Building the ONCE MCP Server

How we built a Model Context Protocol server for AI-powered music distribution.

mcpaiengineering

AI-Powered Genre Mapping

How we built a smart genre resolution system that maps free-text genres to distributor taxonomies.

aiengineeringmetadata

The Validation Agent

An AI agent that sits between conversation and distribution, enforcing Apple Music style rules and catching metadata drift.

aiengineeringinfrastructure