The One About Information Extraction & Decision Systems (ft. OCBC AI Lab)

How do you bridge the gap between what AI retrieves and what business decisions actually need? Explore advanced retrieval systems that go beyond semantic similarity to deliver actionable insights for complex enterprise workflows, featuring real-world applications from OCBC's AI Lab in corporate credit underwriting and SkillsFuture Singapore's approach to workforce intelligence. More About the Sharings OCBC AI Lab Team (Ezekiel Tee & Linus Ng) and OCBC Global Wholesale Banking’s Digital & Transformation Office (Kovan Kok) will be co-sharing on "Custom Retrieval Augmented Generation for Corporate Credit Underwriting" The team will cover how they tackle a Corporate Credit Underwriting use case with their Custom Retrieval Augmented Generation System that utilises a two-phase retrieval and re-ranking architecture to go beyond semantic similarity. They will share in depth how the two-phases are set up to overcome the similarity-utility gap. (Technical Level: 200) Merson Cheong (SkillsFuture Singapore) will share on "Making AI's Impact on Work Measurable: Task Analysis for Workforce Intelligence" This sharing covers SkillsFuture's (SSG) task analysis work and how it helps make AI's impact on work measurable and actionable. Merson will explain the motivation and value of analysing work at the task level—as the practical bridge between jobs and skills—and what this means for workforce skills upgrading and training priorities as technology evolves. The presentation will also cover the technical methodology behind the task analysis, and how the team leverages AI and policy definitions to create a task extractor application. The sharing will include concrete examples and screenshots from the task app, and a brief look at how a task taxonomy can support intelligence and market surveillance. (Technical Level: 200) More About the Speakers Kovan Kok leads the Digital & Transformation Office within OCBC Global Wholesale Banking, where he drives the development of AI use-cases for corporate bankers globally. As chief product owner of the internally developed CRM platform, he oversees its end-to-end build to address RMs' needs across the marketing and client servicing journeys. He holds an MPhil in Economics from the University of Cambridge. Ezekiel Tee leads the Co-Pilot Pod in OCBC's Group Data Office AI Lab, building AI Assistants and agentic workflows for various teams across Consumer Financial Services. He previously developed A.I. Oscar, OCBC's virtual trading assistant with OCBC Securities, drawing on a MSc in Computer Science from Georgia Institute of Technology and experience across marketing and finance. Linus Ng is a Senior Data Scientist in OCBC's Group Data Office AI Lab, where he specialises in architecting and deploying LLM-based systems that support enterprise workflows. His work bridges the gap between research and application to ensure that innovative AI capabilities translate into measurable business impact. He holds a Master of Engineering (M.Eng.) from Nanyang Technological University (NTU), where he conducted research in data science and intelligent systems.Merson Cheong is a Data Scientist at GovTech, where he is forward-deployed to the National Jobs Skills Data Office at SkillsFuture. He specialises in using data science to unlock insights into the labour market, helping to steer national initiatives that connect Singaporeans with the skills needed for the jobs of tomorrow. More About the Series ​AI Wednesdays is Lorong AI’s weekly gathering, bringing together practitioners, researchers and innovators for technical discussions on research insights, product development and engineering practices. ​Get involved: Learn more about Lorong AI | Speaker Sign-up | WhatsApp Community | LinkedIn | X