[SMULA Masterclass] Architecting Private, Auditable AI Systems <Lim How Khang>

Designed for legal practitioners, innovation leads and advisors who must specify or oversee in-house AI deployments, this workshop moves from conceptual talk to systems thinking. Participants analyse cloud versus on-premise trade-offs, map a retrieval augmented generation (RAG) pipeline to their document corpus, and see a “researcher–writer–critic” pattern designed to reduce hallucinations while preserving traceability. An interactive walk through of contrasting AI Verify reports shows how technical tests translate into regulatory assurance. The session ends with construction of an evidence chain demonstrating that accuracy, security and accountability were designed in, not tacked on.

Course Outline

• Architect’s Trade Offs
• RAG Fundamentals
• Multiple Model-Role Workflows
• Evaluation via Automated Tests
• Audit Trail for Governance
• Strategy Clinic and Wrap Up

This masterclass is recommended for participants who have previously completed the instructor’s foundational masterclass or who already use ChatGPT style tools in daily work and now need system design and governance depth.

*Participants should bring a laptop to gain the most out of the hands-on practice session. 

The SMU Law Academy Expert Masterclasses are advanced skills and case-study based programmes where participants will learn from the industry’s top experts in a specific area of practical importance. Typically taught in a face-to-face format, such sessions will help participants to acquire practical in-depth knowhow to aid their development. All the masterclass instructors are specially curated and are recognised as experts in their areas.

SPEAKER