Successfully Scaling AI-Rich Jakarta EE Applications in Production

Successfully Scaling AI-Rich Jakarta EE Applications in Production

Artificial Intelligence is rapidly reshaping enterprise Java but deploying AI-powered applications in production brings its own architectural, performance and cost challenges. Join this JakartaOne session, where Luqman Saeed and Chiara Civardi will unpack what it really takes to run AI workloads reliably and efficiently on Jakarta EE at scale.

In this session, you will learn:

• Why AI-driven applications behave differently from traditional request-response systems and why this makes scaling harder
• Architectural patterns that enable predictable performance for AI features, including async processing, queue-based workflows and fallback strategies
• Testing and validation methods for non-deterministic AI behaviour, including schema validation, golden datasets and drift monitoring
• How to optimise inference cost and performance with techniques such as batching, caching, tiered model selection and prompt efficiency
• Production lessons learned from running real AI-powered Jakarta EE applications including what worked, what failed and what to avoid

This session is designed for developers, architects and engineering leaders who are building or scaling Jakarta EE applications that integrate AI models or natural-language-powered functionality. Gain practical strategies to reduce complexity, improve scalability and achieve predictable results when taking AI from prototype to production.

Secure your place today and learn how to make AI-enhanced Jakarta EE applications production-ready without risking reliability, performance or cost control.