Open to opportunities

Miyuuu Kawatsu

@miyuuukawatsu

Message

I build high-scale LLM platforms, RAG systems, and Go backend infrastructure.

Japan

Message

What I'm looking for

I’ve spent the last few years building backend and AI systems include Frontend at scale, and I’m looking for a role where I can apply that experience to real product problems and keep growing with a strong team.

I've built production LLM platforms at Westons techsolutions, serving over 1M requests per day with p99 latency under 220ms. I architected a FastAPI and Go gateway, a RAG system across 40M documents, and agent workflows that reduced manual processing by about 25%.

I moved ingestion from Python to Go, increasing throughput from 8k to 12k requests per second while reducing infrastructure costs by about 30%. I also built MLOps systems for model registry, rollout, and drift monitoring, cutting regression incidents by around 70%, and mentored four engineers through evaluation-first model review workflows.

Earlier, at Cougar Ridge Group, I delivered BERT and RoBERTa NLP features, semantic search with Sentence Transformers and FAISS, asynchronous inference pipelines, and low-latency Go gRPC services. My development approach is grounded in end-to-end ownership, from prototype to production, alongside the business perspective I developed managing store and back-office operations.

Experience

Work history, roles, and key accomplishments

Senior AI & Backend Engineer

Westons Tech Solutions

Mar 2022 - Jun 2026 (4 years 3 months)

Architected a production LLM platform with FastAPI and Go gateway serving 1M+ requests/day with p99 latency under 220ms. Built RAG systems over 40M documents, agent workflows, and MLOps pipelines, improving relevance by 31% and cutting regression incidents by 70%.