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

Hi, I’m Benjamin Liu. I’m a Staff AI/ML Engineer with about 12 years of experience building production-grade GenAI infrastructure, LLM agent systems, and conversational AI platforms at Google and Amazon. I’ve led initiatives spanning RAG, multi-agent orchestration, ASR/TTS integration, and ML training/serving across Vertex AI, GCP, Amazon Bedrock, Lex, and SageMaker, delivering enterprise-ready tooling and scalable inference services. I enjoy partnering with ML researchers, product, and developer tooling teams to ship impactful features, optimize latency and reliability, and automate complex developer workflows. My goal is to empower teams with robust, observable AI systems that improve task completion, reduce toil, and scale across large organizations.…Hi, I’m Benjamin Liu. I’m a Staff AI/ML Engineer with about 12 years of experience building production-grade GenAI infrastructure, LLM agent systems, and conversational AI platforms at Google and Amazon. I’ve led initiatives spanning RAG, multi-agent orchestration, ASR/TTS integration, and ML training/serving across Vertex AI, GCP, Amazon Bedrock, Lex, and SageMaker, delivering enterprise-ready tooling and scalable inference services. I enjoy partnering with ML researchers, product, and developer tooling teams to ship impactful features, optimize latency and reliability, and automate complex developer workflows. My goal is to empower teams with robust, observable AI systems that improve task completion, reduce toil, and scale across large organizations.

Benjamin Liu

AI Engineer, AI Developer, Data Scientist, +4





Hi, I’m Benjamin Liu. I’m a Staff AI/ML Engineer with about 12 years of experience building production-grade GenAI infrastructure, LLM agent systems, and conversational AI platforms at Google and Amazon. I’ve led initiatives spanning RAG, multi-agent orchestration, ASR/TTS integration, and ML training/serving across Vertex AI, GCP, Amazon Bedrock, Lex, and SageMaker, delivering enterprise-ready tooling and scalable inference services. I enjoy partnering with ML researchers, product, and developer tooling teams to ship impactful features, optimize latency and reliability, and automate complex developer workflows. My goal is to empower teams with robust, observable AI systems that improve task completion, reduce toil, and scale across large organizations.…Hi, I’m Benjamin Liu. I’m a Staff AI/ML Engineer with about 12 years of experience building production-grade GenAI infrastructure, LLM agent systems, and conversational AI platforms at Google and Amazon. I’ve led initiatives spanning RAG, multi-agent orchestration, ASR/TTS integration, and ML training/serving across Vertex AI, GCP, Amazon Bedrock, Lex, and SageMaker, delivering enterprise-ready tooling and scalable inference services. I enjoy partnering with ML researchers, product, and developer tooling teams to ship impactful features, optimize latency and reliability, and automate complex developer workflows. My goal is to empower teams with robust, observable AI systems that improve task completion, reduce toil, and scale across large organizations.

Available to hire

Hi, I’m Benjamin Liu. I’m a Staff AI/ML Engineer with about 12 years of experience building production-grade GenAI infrastructure, LLM agent systems, and conversational AI platforms at Google and Amazon. I’ve led initiatives spanning RAG, multi-agent orchestration, ASR/TTS integration, and ML training/serving across Vertex AI, GCP, Amazon Bedrock, Lex, and SageMaker, delivering enterprise-ready tooling and scalable inference services.

I enjoy partnering with ML researchers, product, and developer tooling teams to ship impactful features, optimize latency and reliability, and automate complex developer workflows. My goal is to empower teams with robust, observable AI systems that improve task completion, reduce toil, and scale across large organizations.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Work Experience

Staff AI/ML Engineer at Google

September 1, 2021 - Present

Staff AI/ML Engineer on Gemini Code Assist Enterprise, building GenAI systems for code understanding, repository-aware assistance, test generation, code review support, and developer workflow automation across enterprise developer environments. Designed multi-agent orchestration across planning, retrieval, code analysis, test generation, and execution paths; built repository-aware RAG pipelines across millions of source files; developed low-latency LLM inference services for prompt construction, retrieval injection, context-window optimization, streaming responses, model fallback, and request routing; deployed ML training, fine-tuning, and evaluation pipelines on Vertex AI to measure code-generation quality, retrieval accuracy, tool-call success, hallucination risk, and regression behavior; implemented model rollout infrastructure with offline evals, A/B tests, canary releases, endpoint monitoring, and rollback workflows.

Staff AI/ML Engineer, L6 at Google

September 1, 2021 - Present

Staff AI/ML engineer on Gemini Code Assist Enterprise, building GenAI systems for code understanding, repository-aware assistance, test generation, code review support, and developer workflow automation across enterprise developer environments. Designed multi-agent orchestration across planning, retrieval, code analysis, test generation, and execution paths; built repository-aware RAG pipelines with embeddings, semantic search, reranking, dependency-aware context assembly, and prompt packaging; developed low-latency LLM inference services with prompt construction, retrieval injection, context-window optimization, streaming responses, model fallback, and request routing. Implemented ML training, fine-tuning, and evaluation pipelines on Vertex AI to measure code-generation quality, retrieval accuracy, tool-call success, hallucination risk, and regression behavior; built model rollout infrastructure with offline evals, A/B tests, canary releases, endpoint monitoring, and rollback workflow

Software Development Engineer II at Amazon

January 1, 2019 - August 1, 2021

Senior backend and AI/ML engineer on AWS conversational AI services, contributing to core Lex runtime services for multi-turn dialogues and voice automation pipelines integrating ASR/TTS, NLU, IVR, Lambda, and event-driven architectures. Built core Lex runtime components for session state, intent routing, slot resolution, fallbacks, and fulfillment handoff; developed voice automation infrastructure spanning Lex, Polly, Transcribe, Lambda, API Gateway, DynamoDB, S3, and Kinesis; integrated ASR transcripts with Transcribe for downstream NLU evaluation; crafted NLU improvement pipelines addressing intent predictions, slot outcomes, and latency metrics.

Software Development Engineer II at Amazon

October 1, 2015 - December 1, 2018

Supported retail personalization and recommendation systems with distributed data pipelines and offline evaluation for customer-facing experiences. Built data pipelines using Hadoop, MapReduce, Hive, and Pig on AWS EMR; developed item-to-item recommendations leveraging co-purchase and co-view signals; led offline evaluation and A/B testing for metrics such as Precision@K, CTR, and revenue-per-session to validate recommendations prior to launch.