The One About Reasoning & Orchestration Systems (ft. OCBC AI Labs)
Explore how enterprise AI systems are moving beyond traditional RAG approaches toward structured information extraction and reasoning workflows. Hear how information is organised, grounded, and orchestrated into systems that generate more reliable and context-aware outputs in practice.*
More About the Sharings
Abhishek (Practitioner, PebbleRoad) will share on “Agentic Orchestration for Customer Support Systems”
Customer support workflows often span multiple systems, manual handoffs, and operational bottlenecks, making reliable AI automation difficult in practice. Abhishek will discuss how orchestration frameworks such as Camunda can coordinate AI agents, business rules, and human decision-making across complex support environments.
Through examples including card eligibility checks, fee waivers, and transaction monitoring, examine how context-aware orchestration layers can support faster resolution times, more reliable automation, and smoother escalation between AI systems and human operators. (Technical Level: 100–200)
David Huang and Ren Xuezhe (Data Scientists, OCBC AI Labs) will share on "AI-Scenario Engine: Evidence-Grounded Outcome Mapping”
Explore the AI Scenario Engine, a reusable framework for AI-assisted decision-making under uncertainty. Using scenario generation as a case study, David and Xuezhe will examine how teams can move from static forecasts toward dynamic, evidence-grounded outcome maps.
Hear how AI systems organize information, contextualize problems, validate assumptions, incorporate human guidance, and map plausible outcomes using Elasticsearch, RAG, LangGraph, and human-in-the-loop governance. Examples will include macro scenario generation, business strategy planning, and risk or policy impact analysis. (Technical Level: 200)
*As the programme is still being finalized, do note that details are subject to change.
More About the Speakers
Abhishek is a practitioner at PebbleRoad who focuses on the design and implementation of intelligent workflows and agentic AI. He specialises in navigating the practical challenges of scaling AI responsibly, moving beyond pilots to build production-grade systems that balance automated reasoning with human-in-the-loop oversight. His recent work includes developing frameworks for agentic orchestration that enable organisations to embed intelligence into complex, high-stakes operational processes.
David Huang is a data scientist specialising in agentic AI solutions for the banking industry, with end-to-end expertise spanning data pipeline engineering, Retrieval-Augmented Generation, semantic search, prompt tuning, graph-based workflow design, and production deployment using frameworks such as LangGraph. As a former BCG consultant, he has worked closely with senior stakeholders to co-design multi-agent workflows that accelerate idea generation, enhance decision quality, and translate complex business challenges into production-ready AI applications. Earlier in his career at Shopee, he applied machine learning and data engineering using Spark, SQL, and MLOps practices to improve marketing campaign conversion rates through user targeting.
Ren Xuezhe is a data scientist specializing in the design and deployment of agentic AI solutions across the finance industry, with a strong focus on fintech. He works closely with senior stakeholders and domain experts to architect multi-agent systems that drive ideation, enhance decision-making, and transform complex challenges into scalable, production-grade AI applications.
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