I build distributed systems and lead the teams that run them at scale. The talks below, spanning more than a decade, cover the work directly: database internals, observability infrastructure, and what it takes to design, ship, and operate systems that other engineers depend on.
Conferences
Unified Management Plane for OpenTelemetry
Google Cloud Next 2026 · Las Vegas, NV
Co-presented with Keith Chen of Google Product Management on how Google Cloud gives customers control over their telemetry before it is stored. The talk covered the design of the Telemetry Router — a managed OpenTelemetry entry point that filters, transforms, aggregates, and routes metrics, logs, and traces to Cloud Observability or to third-party backends. We showed how server-side OpenTelemetry control can reduce client-side complexity, give enterprises more control over telemetry cost, and preserve interoperability across observability backends..
Build Your Own Distributed System Using Go
GopherCon 2023 · San Diego, CA
A hands-on walkthrough of building a distributed key-value store using Go and the Hashicorp Raft library. The talk covers distributed consensus, snapshotting, log compaction, and the practical differences between distributing a simple key-value store versus a more complex storage engine like BoltDB.
Observing Observability
DevOpsDays Galway 2019 · Galway, Ireland
A version of my Monitorama 2019 talk, delivered at DevOpsDays.
Observing Observability
Monitorama 2019 · Baltimore, MD
Why observability and monitoring systems struggle to meet their goals, and why they are so hard to build. Drawing on my experience at Loggly, InfluxData, and Google Cloud, I examine the recurring patterns and mistakes teams make when building these systems at scale.
Infrastructure at Scale: Apache Kafka, Apache Storm & Elasticsearch
AWS re:Invent 2013· Las Vegas, NV
Co-presented with Jim Nisbet (CTO, Loggly). A presentation of how Loggly built its second-generation multi-tenant, geo-distributed, real-time log management system on AWS, using Kafka, Storm, and Elasticsearch. This talk was one of the first public accounts of using Kafka as a durable ingestion layer in front of Elasticsearch for large-scale log aggregation. Both Kafka and Elasticsearch were pre-1.0 and the design showcased a pattern that later became common across log analytics systems.
Academic
rqlite: The Distributed Database Built on Raft and SQLite
Carnegie Mellon Database Group – Vaccination Database Tech Talks, 2021
Invited by the CMU Database Group to present on rqlite’s goals, design, Raft consensus integration, SQLite embedding, testing strategy, and real-world applications.
Distributed Systems with rqlite
ACM Pittsburgh – 2018 Series
An introduction to distributed systems, the problems they solve, and how rqlite uses Raft to replicate SQLite.
Regional
Ekanite: The Syslog Server with Built-in Search
San Francisco Go Meetup
A talk on Ekanite, a syslog server I built in Go with integrated full-text search using the Bleve library.
InfluxDB and the Raft consensus protocol
San Francisco Database Meetup
As a core member of the InfluxDB development team I gave a presentation on InfluxDB internals and its use of the Raft consensus protocol.
Talking Distributed Systems at Hacker Nights
Hacker Nights
A talk on rqlite and distributed systems to the New York-based development group.
Podcasts
rqlite and Open Source development
Contributor Podcast · hosted by Eric Anderson, Scale Venture Partners
A wide-ranging discussion on rqlite, its design, the open-source development model, and the economics of building and maintaining open-source software.
Cloud Logging on Google Cloud Platform
Google Cloud Platform Podcast · with Reed Taylor
A discussion of Cloud Logging, GCP’s centralised logging product, its architecture, and recent feature additions.
Audit Logging on Google Cloud Platform
Google Cloud Platform Podcast · with Oscar Guerrero
A conversation on GCP’s Audit Logging capabilities and their role in security and compliance.
Writing on Speaking
I have also written about the craft of conference speaking itself:
Interested in having me speak? Get in touch.