I’m Darby Huye. I live in Boston, where I research observability in the cloud. I'm currently open to new opportunities!

I'm graduating with my PhD from Tufts University in May 2026. I researched in the D.O.C.C. Lab at Tufts studying under the guidance of Professor Raja Sambasivan. My research interests involve leveraging distributed tracing for debugging performance problems in cloud-based systems.

The first portion of my Ph.D. research was centered on understanding the buzzword microservices. I (along with my colleagues) conducted a user-study interviewing folks in industry and academia about their experiences with microservice-based applications. We found that the assumptions around this buzzword and the open-source academic testbed applications are far too limited to enable robust research in this space. This lead me to an internship at Meta, where I had the opportunity to analyze aspects of their microservice applications. My findings were published at ATC '23 and a portion of the data was released on github.

The second portion of my Ph.D. was investigating: 'how can we capture meaningful observability data in our systems?' Prior research in this space assumes perfect observability (or ignores the possibility of imperfect data), which is far from the reality at many large organizations. I investigated the degree to which data is imperfect in open source traces (from Alibaba, Meta, and Uber) and designed a new primitive to bridge the gap in traces when data is lost.

I'm currently looking for new work opportunities! I'm interested in working in the observability space, particularly around agentic-ai systems, automating observability decisions for developers (e.g., what data should we collect, how we should collect it, how can we show our observability data is useful for our specific needs, etc.), and helping build new tools for developers to use in their daily workflows. Please reach out over email if you're interested in chatting.