Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

Hi, I’m Yang Liu, a Full Stack AI Engineer based in Sydney, Australia with 7+ years of experience spanning React, Node.js, and Java. I specialize in LLM integration, agent orchestration (LangChain, Crew AI), and building end-to-end multimodal AI pipelines that scale for production. I design scalable full‑stack systems, implement real-time dashboards, and craft intelligent, agentic experiences using RAG, prompt engineering, and vector databases. I’ve built orchestration layers, backend APIs, and observability tooling to monitor performance across high-traffic services and continually improve model behavior.…Hi, I’m Yang Liu, a Full Stack AI Engineer based in Sydney, Australia with 7+ years of experience spanning React, Node.js, and Java. I specialize in LLM integration, agent orchestration (LangChain, Crew AI), and building end-to-end multimodal AI pipelines that scale for production. I design scalable full‑stack systems, implement real-time dashboards, and craft intelligent, agentic experiences using RAG, prompt engineering, and vector databases. I’ve built orchestration layers, backend APIs, and observability tooling to monitor performance across high-traffic services and continually improve model behavior.

Yang Liu

Full Stack Developer, AI Engineer, Data Scientist, +2





Hi, I’m Yang Liu, a Full Stack AI Engineer based in Sydney, Australia with 7+ years of experience spanning React, Node.js, and Java. I specialize in LLM integration, agent orchestration (LangChain, Crew AI), and building end-to-end multimodal AI pipelines that scale for production. I design scalable full‑stack systems, implement real-time dashboards, and craft intelligent, agentic experiences using RAG, prompt engineering, and vector databases. I’ve built orchestration layers, backend APIs, and observability tooling to monitor performance across high-traffic services and continually improve model behavior.…Hi, I’m Yang Liu, a Full Stack AI Engineer based in Sydney, Australia with 7+ years of experience spanning React, Node.js, and Java. I specialize in LLM integration, agent orchestration (LangChain, Crew AI), and building end-to-end multimodal AI pipelines that scale for production. I design scalable full‑stack systems, implement real-time dashboards, and craft intelligent, agentic experiences using RAG, prompt engineering, and vector databases. I’ve built orchestration layers, backend APIs, and observability tooling to monitor performance across high-traffic services and continually improve model behavior.

Available to hire

Hi, I’m Yang Liu, a Full Stack AI Engineer based in Sydney, Australia with 7+ years of experience spanning React, Node.js, and Java. I specialize in LLM integration, agent orchestration (LangChain, Crew AI), and building end-to-end multimodal AI pipelines that scale for production.

I design scalable full‑stack systems, implement real-time dashboards, and craft intelligent, agentic experiences using RAG, prompt engineering, and vector databases. I’ve built orchestration layers, backend APIs, and observability tooling to monitor performance across high-traffic services and continually improve model behavior.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Language

English

Fluent

Work Experience

Full Stack AI Engineer at TikTok

January 1, 2024 - Present

Developed a full‑stack multimodal LLM content moderation system (GPT-4V and internal models) to automatically detect inappropriate video, image, and text content, reducing manual review volume by an estimated 35%. Built an agent orchestration layer using LangChain and custom Python services to coordinate vision, language, and safety classifiers across 15 content categories. Created a real-time moderation dashboard (React + TypeScript) for trust & safety teams to review flagged content, view LLM reasoning traces, and provide feedback for continuous improvement. Designed high‑throughput backend APIs in FastAPI and Node.js with Kafka for streaming video data, supporting 50k+ moderation requests per day with sub-second latency. Implemented a vector database (Pinecone) for few-shot example retrieval, enabling adaptation to new harmful patterns without full retraining.

Full Stack AI Developer at ByteDance (Shanghai)

September 1, 2019 - January 1, 2024

Architected and developed a core component of Eino, ByteDance’s internal LLM application orchestration framework, built in Go and used by hundreds of services including Doubao and TikTok. Designed and implemented the framework’s component abstraction layer with standard Go interfaces for AI building blocks (ChatModel, Tool, Retriever, Lambda), enabling flexible and reliable integration of diverse models and tools. Built the graph-based orchestration engine for production-grade AI workflows using directed graphs (DAGs) to support flexible, scalable agent systems. Developed tooling for observability (logs, traces) to enable product teams to trace execution, log steps, and monitor performance across AI applications. Collaborated with internal teams to gather feedback and refine APIs to ensure production-hardening for high-traffic services.

Full Stack Intern at Wish

May 1, 2019 - August 1, 2019

Developed responsive frontend components using React and Redux for the merchant dashboard, improving page load time by 20%. Built Node.js and Express REST APIs to manage product inventory and orders, integrated with MongoDB. Implemented unit tests with Jest and participated in code reviews following Agile methodology.