About Me


I'm Takao Shibamoto, a software engineer. Most of my work has been on the parts of an AI product that don't get much press — the pipelines that move data around, the control planes that keep services running, the data layers that have to keep working when volume triples overnight. That tends to be the bottleneck when enterprises try to adopt AI, and it's where I've spent most of the last seven years.

Currently at Microsoft, building large-scale distributed control-plane infrastructure for enterprise AI products. Before that: Amazon on last-mile delivery, Everlywell on at-home lab testing through the pandemic, and PayPal as an intern during college on corporate firewall automation.

I work mostly on distributed systems and data infrastructure, but I've shipped plenty outside that lane too — full-stack web apps, computer-vision projects, automation tools, and a handful of hackathon prototypes. If the problem is interesting I'll figure out the stack.

Outside work, I enjoy traveling, music, and following the markets.


Selected writing:
Open source & research:
  • Floe — an open-source engine for building dynamic tables on Apache Iceberg. Tables are defined in Python; Floe handles the orchestration, event-driven updates, and cloud-agnostic deployment. GitHub.
  • Primer Server — a web tool I built as an undergrad in the Vollbrecht Lab at Iowa State to help biology researchers design PCR primers. Open-sourced from day one. GitHub · paper (Journal of Student Research, 2018).
Experience:
  • Software Engineer @ Microsoft (2026 – present) — large-scale distributed control-plane infrastructure for commercial enterprise AI products.
  • Software Engineer @ Amazon (2021 – 2026) — event-driven big-data systems behind last-mile delivery performance, at worldwide scale.
  • Software Engineer @ Everlywell (2020 – 2021) — backend systems for the lab and fulfillment operations behind at-home testing, during the pandemic.
  • Software Engineer Intern @ PayPal (2018 & 2019) — Python tools and internal web applications for corporate firewall automation.
Skills:
  • Languages: Python, Java, TypeScript, Scala, Ruby
  • Data systems: SQL, NoSQL, distributed processing (Spark), ETL, BI & visualization
  • Infrastructure: containerization (Docker / Kubernetes), Infrastructure as Code, serverless, monitoring & observability
  • Distributed systems: event-driven architecture, message queues, stream processing, object storage
  • AI: LLM integration (LangChain), computer vision (OpenCV), model-serving infrastructure, retrieval-augmented generation (RAG)
Education:
Online presence:
Contact:
  • takao at takasoft.io