Learn how Braze AI and Liquid power dynamic messaging.
About this course
Discover how to use Liquid as the connective tissue between your customer data and Braze’s AI features. In this hands-on workshop, led by trainer @SangK, you’ll learn how Braze AI handles the intelligence while Liquid manages delivery, enabling you to build deeply personalized, automated customer journeys.
The session begins with a fast-paced Liquid primer covering core syntax—objects, tags, and filters—along with attributes, logic, and proper fallbacks. From there, the training dives into the four primary places where Liquid and Braze AI intersect:
- Braze AI Operator: Learn to prompt Braze’s AI agent to generate custom Liquid code, refine responses, and audit generated syntax for proper edge cases and abort logic.
- AI Item Recommendations: Connect catalog data to machine learning models (personalized, trending, most popular, most recent) and pull zero-indexed recommendation objects directly into your message body.
- Predictive Suite: Leverage predictive churn and event scores as targeting filters, using Liquid conditionals to vary content and promotional tiers based on customer risk profiles.
- Agent Console: Pass user attributes into autonomous AI agents using Liquid syntax, then capture and render the agent's output as dynamic Canvas context variables.
By the end of this training, you’ll understand how to stack these tools together within a single Canvas and possess the practical knowledge required to complete the Braze Liquid Essential Skills Badge.
After completing this course, you will be able to:
- Understand objects, output markers, tags, filters — the core building blocks.
- Generate Liquid with Operator™ — and know what to verify first.
- Surface AI recommendations in a message via product_recommendation.
- Segment by churn/event score, then tailor content per tier.
- Distinguish which AI touches content vs. delivery — and Liquid.
- Combine multiple BrazeAI features in one Canvas, tied together by Liquid.
Can't make it? Watch the replay here.
