Yuki Nakamura
Fullstack & AI Engineer specializing in agentic AI systems, RAG pipelines, and LLM applications.
Turning AI prototypes into scalable, cost-optimized products with FastAPI, Next.js & the modern AI stack.
Yuki Nakamura is a Fullstack & AI Engineer with hands-on production experience building agentic AI systems using Google's Agent Development Kit (ADK), Vertex AI Agent Engine, and modern RAG architectures. A Japan-based engineer with Tokyo industry experience, he deploys end-to-end LLM applications combining FastAPI, Next.js, and Model Context Protocol (MCP) tooling — and is skilled at turning AI prototypes into production-ready, cost-optimized cloud microservices. Available immediately for remote AI Engineering roles or visa-sponsored positions.
Yuki Nakamura
Design and deployment of multi-agent systems on Google Vertex AI Agent Engine and the Agent Development Kit (ADK), including secure authentication and agent-as-a-tool composition. From task decomposition and tool selection to escalation logic and evaluation pipelines, I build agents that run reliably in production.
Learn More
Production-grade backends with FastAPI, Flask, and Node.js — async services, robust error handling, state persistence, and monitoring. I connect systems to PostgreSQL, MongoDB, Redis, and external APIs, and build standalone multimodal services for audio, image, and video processing.
Learn More
Production RAG pipelines using Google's RAG Engine — automated corpus management, chunking, embedding, and custom retrieval tools. With vector DBs (Qdrant, Pinecone) and evaluation pipelines (Ragas), I've reached 92%+ factual accuracy on real production query traffic.
Learn More
Workflow automation with n8n, Zapier, and make.com, plus Selenium, webhooks, and Model Context Protocol (MCP) tooling. I wire LLMs into Microsoft Graph, SharePoint, Telegram, and self-hosted infrastructure to remove manual steps and cut operational time.
Learn More
CompTIA CySA+ certified with Splunk Core and Cisco Network Defense credentials. I apply secure coding practices, vulnerability assessment, and monitoring to AI and backend systems — building trust through safe authentication, data handling, and observability.
Learn More
Deploying cost-optimized microservices on Google Cloud (Vertex AI, Cloud Run) with Docker, Kubernetes, and CI/CD. I also deliver self-hosted infrastructure — e.g. a Nextcloud platform with automated backup and disaster recovery that cut storage spend by an estimated 60–70%.
Learn More
I start by understanding the problem and rapidly prototyping an AI solution — validating prompts, retrieval, and agent design before scaling.
I engineer the full stack — FastAPI backends, Next.js frontends, RAG pipelines, and agent orchestration — with testing, evaluation, and monitoring built in.
I ship to Google Cloud as cost-optimized microservices, then measure accuracy and latency in production and tune for reliability and spend.
Fullstack
(React/Next.js, TypeScript, Node.js, Python)
AI & LLMs
(ADK, Vertex AI, RAG, MCP, FastAPI)
Yuki Nakamura
Get in touch
Fullstack & AI Engineer
Available immediately for remote AI Engineering roles or visa-sponsored positions.
