August 22, 2026•1 min read•from InfoQ

Presentation: SafeChat: Building AI-Powered Safety Systems at Scale in a Real-Time Marketplace

Bruna Pereira explains how DoorDash built a content-agnostic AI moderation platform. She covers replacing costly LLM-only pipelines with a hybrid pattern: using fast internal models to filter obvious cases, LLM multi-axis scoring for nuanced decisions, and no-code workflows with backtesting. Discover how this architectural pattern cut safety incidents while scaling to millions of daily messages.

By Bruna Pereira

Want to read more?

Check out the full article on the original site

View original article→

Tagged with

#AI Moderation

#Content-Agnostic

#LLM

#Large Language Models

#Hybrid Pattern

#Real-Time Marketplace

#Safety Systems

#Safety Incidents

#Scaling

#DoorDash

#Internal Models

#Multi-Axis Scoring

#No-Code Workflows

#Backtesting

#Architecture

#Content Filtering

#Pipelines

#AI-Powered

#Message Moderation

#Bruna Pereira