AI Builders

Chairpersons welcome and opening remarks

As the automotive industry shifts from one-time vehicle sales to long-term, service-driven relationships, AI and big data are becoming critical enablers of transformation. This session explores how Hyundai Motor is centralizing customer, digital, sales, and physical data to build a unified intelligence ecosystem. By leveraging predictive analytics and AI-powered segmentation, Hyundai is unlocking personalised customer journeys, optimizing digital touchpoints, and enabling new subscription-based services, even before the purchase phase. Discover how data-driven insights are reshaping customer engagement, increasing lifetime value, and redefining mobility experiences.

AI agents are rapidly evolving from simple assistants into systems capable of planning, reasoning and executing complex workflows. Building reliable agentic applications requires more than powerful models—it demands orchestration, memory, tool integration and robust evaluation. 

Discover how modern engineering teams are designing scalable agentic systems capable of delivering reliable outcomes while maintaining security, governance and developer productivity. 

Retrieval-Augmented Generation has quickly become a foundation for enterprise AI, but delivering accurate, trustworthy responses requires far more than connecting a vector database to a language model. 

Learn how engineering teams are improving retrieval quality, managing enterprise knowledge, reducing hallucinations and building scalable architectures that support production-grade AI assistants. 

Deploying an AI application is only the beginning. Engineering teams now require visibility into latency, drift, hallucinations, cost, quality and user behavior to maintain reliable production systems. 

This session explores emerging observability practices that enable teams to continuously monitor, evaluate and optimize AI applications throughout their lifecycle. 

Shipping production AI requires balancing performance, reliability, security and developer velocity. This panel explores how engineering teams are redesigning software architectures to support foundation models, agentic workflows and increasingly complex AI-powered applications. 

Topics include: 

• AI-native software architecture  
• Model orchestration and routing
• Context engineering and memory  
• Testing and evaluation  
• Scaling AI engineering teams  

The AI ecosystem is expanding rapidly, giving developers more choice than ever before. Large proprietary models, open-weight alternatives and smaller task-specific models each offer different advantages depending on cost, latency, privacy and performance requirements. 

Learn how engineering teams are selecting, combining and managing multiple models to optimize both technical performance and commercial outcomes. 

As AI workloads scale, engineering teams are increasingly focused on reducing latency, managing GPU utilization and controlling inference costs without sacrificing quality. 

Discover practical optimization techniques covering model compression, batching, caching, hardware acceleration and efficient deployment architectures. 

Traditional software testing cannot adequately evaluate modern AI systems. Measuring quality now requires continuous evaluation of reasoning, factual accuracy, safety, robustness and business performance. 

This session explores emerging evaluation frameworks, automated benchmarking and human feedback techniques that help engineering teams confidently deploy AI into production. 

The AI engineering landscape is evolving faster than any other area of software development. Agentic systems, multimodal applications, AI-native operating systems and increasingly autonomous development tools are changing how software will be designed over the next five years. 

This closing session explores the technologies, frameworks and engineering practices developers should be investing in today to stay ahead of the next generation of AI innovation. 

Moving AI from experimentation into production introduces technical challenges rarely encountered during development. Infrastructure complexity, governance requirements, deployment pipelines and integration with existing enterprise systems all become critical factors in long-term success. 

Hear from engineering leaders who have successfully deployed AI at scale as they share practical lessons, unexpected challenges and the technical decisions that made the biggest difference. 

Topics include: 

• AI platform engineering  
• Deployment automation  
• Security by design  
• Cost optimization  
• Engineering for resilience