Liz Liu

Data Scientist.

I work at the intersection of GenAI applications, rigorous experimentation, and business impact — turning complex models and data into decisions that matter.

Currently at Iress, Singapore. Previously at Commonwealth Bank, ByteDance, and ING.

Python SQL RAG LangGraph PyTorch GCP

View Writing Get in Touch →

Most AI projects fail not because of the models, but because nobody asked the right question. I care about the full loop — understanding the problem, running the experiment rigorously, and making sure the result actually lands with the people who need to act on it.

FEATURED PROJECTS

Iress

Production RAG System

Designed and deployed multi-stage RAG with chunking strategies, hybrid retrieval, and reranking to reduce hallucination in production.

RAGPythonHybrid RetrievalReranking

Iress

GenAI Evaluation Harness

Built an automated eval pipeline covering test-case generation, multi-dimensional scoring, regression detection, and human-in-the-loop feedback.

EvaluationDeepEvalHITL

Iress

LLM Meeting Summarizer

LLM-powered system that extracts key client information from meetings and generates structured summaries, cutting manual post-meeting workload.

LLMAutomationBusiness Impact

Commonwealth Bank

Customer Churn Propensity Model

Led full model lifecycle — training, deployment, monitoring — for churn prediction; converted batch model to real-time inference service.

PyTorchMLOpsReal-time Inference

ByteDance

Advertising Experimentation

Used causal inference and A/B testing to measure the impact of product and algorithm changes on advertising revenue and user engagement.

Causal InferenceA/B TestingRevenue Impact

LATEST WRITINGS

Everyone's Misreading Karpathy's LLM-Wiki Gist

It's not about replacing RAG with Markdown. It's about turning an LLM from a one-shot answering machine into a compounding editor of a growing knowledge asset.

April 15, 2026

View All Posts →

LET'S CONNECT

lizliu627@gmail.com LinkedIn → GitHub (coming soon)