Data & AI Architecture Sydney schedule

In this 10-minute networking session, the goal is to connect with three new people. Enjoy the opportunity to expand your network!

• How should enterprise AI agents be designed for scale and reliability?
• When does a multi-agent architecture outperform a single-agent approach?
• Where should human oversight be embedded in agent workflows?
• What lessons have organisations learned from production deployments?

• What does reliability engineering mean for data teams?
• How should organisations define and measure data SLAs and SLOs?
• Which monitoring practices best detect issues before users do?
• How can teams respond to and learn from data incidents? 

• What architecture decisions delivered the greatest value?
• Which assumptions proved wrong once solutions reached production?
• Where did organisations encounter unexpected complexity and technical debt?
• What would architects do differently if starting again today?

Speakers
Vitus Chu Head of Data Strategy & Data Architecture Acenda 

When Resimac compared the pipelines behind its reporting, analytics, ML and GenAI workloads, it found far more overlap than expected. This session walks through the resulting architecture, illustrated through Resimac’s AI-driven business strategy, shares how practitioner feedback reshaped it, and closes with a golden path for bringing more GenAI workloads into the governed enterprise platform.

• Where do BI, ML and GenAI pipelines genuinely overlap, and where do they diverge?
• What changes when data quality moves from batch aggregates to retrieval-time answers for GenAI?
• How should AI agents use existing ML models, and what are the failure modes?
• How can architects create a golden path that makes the governed platform the default for GenAI?

• Are knowledge graphs becoming essential for enterprise AI?
• How does a semantic layer improve consistency and trust?
• What role should business metadata play in modern architecture?
• How can organisations connect knowledge, context and AI outcomes? 

• What separates successful data products from failed ones?
• Who should own and manage data products?
• How much standardisation is needed across domains?
• How can organisations improve data product discovery and adoption? 

• Why has AI renewed interest in master data management?
• How do organisations create trusted golden records at scale?
• Can modern entity resolution solve long-standing data quality challenges?
• What does a future-ready MDM architecture look like?

• Why is metadata becoming the foundation of modern data platforms?
• How can active metadata automate governance and operations?
• What does effective lineage look like at enterprise scale?
• Can metadata become the control plane for data and AI? 

• Has data mesh delivered on its promise?
• What role does data fabric play in modern architecture?
• Are data products succeeding as a delivery model?
• Should organisations centralise, federate or adopt a hybrid approach? 

• Which integration patterns are best suited to modern enterprises?
• Where do APIs, events and streaming each add value?
• How should architects define and manage domain boundaries?
• What does an integrated data product ecosystem look like? 

• Which architectural patterns break down when supporting AI workloads?
• How should transactional, analytical and AI data architectures converge?
• What new architectural components are becoming essential?
• How do architects balance structured, unstructured and semantic data?