AI & Analytics leader building production-shaped agentic AI systems — currently Analytics Manager at Accenture, based in Singapore.
What I work on: enterprise AI/analytics products end-to-end, from opportunity identification through MLOps deployment — spanning core analytics disciplines (customer lifetime value, marketing mix modelling, multi-touch attribution, pricing & elasticity, churn/segmentation, next-best-action) as well as the GenAI/agentic layer on top. The client-facing analytics work is enterprise/proprietary, not open source; the repos below are where I open-source the underlying agent-engineering patterns instead.
Featured projects
- Agentic-AI-Compliance-Copilot — LangGraph multi-agent system that checks an AI system for compliance against NIST AI RMF, IMDA's Model AI Governance Frameworks, and ISO/IEC 42001, with every verdict re-verified against retrieved source text by a citation guardrail.
- AutoCare-Service-Multi-Agent-Support — Hierarchical LangGraph multi-agent assistant for an auto parts & service center, with cross-customer data isolation enforced by graph wiring.
- kyc-pipeline — AI-assisted document processing for corporate bank KYC onboarding: extraction, ownership-chain resolution, sanctions screening, human-reviewed recommendations with full provenance.
Stack: Python · LangGraph · LangChain · RAG · MCP · FastAPI · Docker · GCP / AWS / Snowflake · Power BI / Tableau