EntityMap
Duncan Kuo

Duncan Kuo

Data & AI Engineering

I build data platforms and AI engineering that ship — 8+ years driving 50%+ infrastructure cost cuts, reclaiming 80+ engineering hours weekly, and moving AI agents from prototype to production.

LinkedIn Contact

What I build here

Data pipelines & backend

Ingestion, storage, and the backend services behind the analyzers — keeping the tools fast and the data clean.

AI engineering

LLM agent design and MCP-based tool wiring that powers the fan-out, entity, and citation analyzers under the hood.

Architecture & tech selection

Knowing when to add an LLM and when not to, when batch beats streaming, where to put the embeddings, what to cache — the unglamorous decisions that decide whether the tools stay fast and cheap as traffic grows.

DataOps & governance

Observability, lineage, data quality, and access controls so results stay trustworthy as the toolset grows.

Background

I've spent the last several years building data platforms end to end — from analytics warehouses and BI to real-time streaming, ML pipelines, and the data governance work that keeps everything trustworthy at scale. The same systems that ship business insight every day are the ones that have to stay correct, observable, and fast.

More recently my focus has shifted toward AI engineering: LLM agents, MCP-based tool wiring, retrieval-augmented systems, and the orchestration layer that ties generative models back to the structured data they need to be useful. It's a natural extension of platform work — the failure modes are different, but the disciplines (clear interfaces, observability, evaluation) are the same.

EntityMap is where those two worlds meet. I'm the engineering half of the site — the pipelines, AI agents, and infrastructure behind the analyzers. Solid infrastructure underneath, AI on top, and a deliberate effort to keep the tools honest about what they actually measure.

Expertise

Data ArchitectureAI Engineering (LLM Agents, MCP)Cloud (AWS / GCP)Data LakehouseETL (DBT, Airflow)Real-time Streaming (Kafka, Spark)Kubernetes (GKE)DataOps & CI/CDData GovernanceBI & Analytics

Speaking