ziwen.ai

Bringing AI models
into production.

Ziwen Cao Staff Software Engineer

At ByteDance / TikTok, I build AI infrastructure for model deployment, workflow orchestration, and GPU scheduling.

Explore my AI work Connect on LinkedIn (opens in a new tab)
Ziwen Cao outdoors in the mountains
Ziwen (Chase) CaoSan Mateo, California

Experience since February 2016 in software engineering

ByteDance / TikTok Previously Meta & Airbnb

AI systems, in practice.

Model serving, workflow orchestration, and GPU scheduling.

Generative models.
Production infrastructure.

ByteDance / TikTok · Current work

AI infrastructure on the Large Vision Model Engineering team, with ownership of workflow orchestration and scheduling for AIGC serving.

Role & experience

Model deployment & serving

Deploying and serving Seedance and Seedream, alongside open-weight models including Qwen and MiniMax H3.

vLLM CUDA ComfyUI

Orchestration & GPU utilization

Workflow orchestration and scheduling for an asynchronous AIGC serving framework, with a focus on improving GPU utilization.

Architecture & adoption

Making architecture decisions, coordinating work across teams, and working with users and leadership on adoption and project funding.

Built on distributed systems.

A career in backend infrastructure, real-time systems, and AI.

ByteDance / TikTok

— Present

Staff Software Engineer

Large Vision Model Engineering

Own AI infrastructure for AIGC serving, workflow orchestration, scheduling, and model deployment. Mentor engineers and help teams adopt coding-agent workflows, including automating repetitive engineering work.

AI infrastructure · Model serving · Technical leadership

Meta

—

Senior Software Engineer

Reality Labs Payment Risk

Led a cross-functional CVV initiative for Reality Labs purchases. Built payment risk systems for Reality Labs’ app store, including fraud detection, real-time anomaly detection, and rate limiting.

Applied ML · Payment risk · Real-time systems

Airbnb

—

Software Engineer → Senior Software Engineer

Architected foundational SOA services for Airbnb’s move from a monolith to microservices. Built a Kafka-based pub-sub system and managed critical services including Dynamic Configuration.

Led ChatGPT integration into the customer support chatbot platform. Also worked across search, advertising, and recommendations.

Distributed systems · Kafka / SOA · LLM integration

Earlier experience Zenefits & Amazon Web Services

Zenefits

—

Full Stack Software Engineer

Helped build and launch an internal insurance platform.

Amazon Web Services

—

Software Development Engineer Intern · Amazon EMR

Designed and built a web-based tool to automate log analysis for Amazon EMR.

Always learning.
Sometimes skiing.

I’ve spent most of my career building backend and distributed systems. These days, I work on the infrastructure that brings AI models into production.

Outside work, I study LLM inference systems such as vLLM and SGLang, and experiment with operator fusion in Triton. I also love skiing, including black diamond runs.

Based in
San Mateo, California
Education
USC · M.S. in Computer Science

Let’s talk AI infrastructure.

For opportunities in AI infrastructure, model serving, and the systems behind them, connect with me on LinkedIn.

Connect on LinkedIn (opens in a new tab)