The One About Agentic Workflows (ft. OCBC AI Lab)
How do AI agents coordinate and adapt across multi-step workflows? Explore the possibilities of agentic systems in real-world business processes, showing how agents can be programmed and deployed to manage complex workflows in practice.
More About the Sharings
Lim Jia Yi (Senior Data Scientist, OCBC AI Lab) will share on “Agentic Workflow for Better Customer Experience”
Managing bonus interest appeals of client enquiries requires verifying complex eligibility criteria across multiple banking products and transaction patterns, a traditionally manual and time-intensive process. Jia Yi will share how OCBC has harnessed LangGraph to orchestrate a workflow for handling these appeals - by extracting the required information, routing it to different nodes based on the extracted data, dealing with edge cases, updating shared state, and synthesising the final outputs. (Technical Level: 100)
Praveen Govindaraj (Lead AI Engineer) will share on "Neam: A Compiled DSL for AI Agents"
Praveen will explore how AI agents can be orchestrated through a compiled domain-specific language (DSL). He'll demonstrate how Neam expresses Claw agents (long-running) and Forge agents (Plan → Execute → Verify) in ~30 lines of code, showing how agents can spawn others, execute in DAG order, and maintain persistent semantic memory. He’ll show how Neam’s primitives—skills, guards, sessions, channels, and multi-agent coordination—help surface configuration and runtime issues during compilation. The language compiles to bytecode and can generate deployment artifacts for Docker, Kubernetes, and AWS Lambda. (Technical Level: 200-300)
Jeffrey Lim (AI Governance and Safety Manager, IMDA) will share on “Model AI Governance Framework for Agentic AI”
As AI agents become more autonomous and capable of taking real-world actions, new governance challenges emerge around accountability, risk management, and human oversight. Jeffrey will present IMDA’s newly released Model AI Governance Framework for Agentic AI, which provides organisations with structured guidance for responsible agent deployment. Hear more on the framework's four key dimensions: assessing and bounding risks upfront through appropriate use case selection and agent limits, maintaining meaningful human accountability across complex multi-stakeholder environments, implementing technical controls throughout the agent lifecycle, and enabling end-user responsibility through transparency and training. Learn practical approaches to managing agentic risks while unlocking their transformative potential. (Technical Level: 100)
More About the Speakers
Jia Yi is a Senior Data Scientist in the Monetisation team at OCBC Bank’s Artificial Intelligence Lab. She previously worked as an NLP Engineer at Lex Machina, a legal technology firm spun out of Stanford University, where she developed natural language processing solutions for their best-in-class legal analytics platform. Jia Yi holds a Master's degree in Computer Science specialising in Machine Learning from the Georgia Institute of Technology.
Praveen Govindaraj is a Lead AI Engineer at a telecommunications company in Singapore, driving business value through strategic and responsible AI initiatives. He specializes in designing and deploying LLM and agent-based workflows across financial services and telecom. Praveen is the creator of Neam, an AI-native programming language for agentic systems that integrates agents, reasoning, and orchestration directly into code. With over a decade of experience, he has contributed to projects at DBS Bank and OCBC Bank, including customer engagement, AML/CFT, credit risk, and enterprise data platforms. Outside of work, he enjoys fitness, soccer, and music.
Jeffrey Lim is a member of IMDA's AI Governance and Safety team, where he works with industry and public agency stakeholders to develop Singapore's AI governance standards, guidelines and policies. He specialises in governance frameworks for emerging AI technologies, particularly AI agents. Jeffrey began his career as a software engineer in the interactive media industry before transitioning to the public sector, where he has worked on technology governance and policy across diverse areas including law enforcement and media regulation.
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