krellis.org

Tim Wilde

Systems, software and network engineering triple-threat

Download PDF

How I work

  • 1 / 4

    Run towards the hard problem

    I specialize in deep-dive debugging of production systems and the code that runs on them, and I’m happiest jumping into something unfamiliar when it matters.

  • 2 / 4

    Automate the toil, reuse what exists

    Reuse existing and off-the-shelf solutions where possible, and automate repetitive work, so teams can spend their time on engineering rather than operations. I’m interested in where AI can help with that, with care about what it is allowed to change in production.

  • 3 / 4

    Let SLOs decide where effort goes

    Carefully crafted SLOs balance business needs, operational realities and competing goals like reliability and velocity, and they say what to focus on versus what can safely wait.

  • 4 / 4

    Leave the practices behind

    I set best practices on the teams I join, through example as much as through documentation and policy-setting, so good operational habits outlast any one incident.

Experience

  1. Sep 2012 – Present

    Google

    Cambridge, MA

    • Staff Site Reliability Engineer Oct 2022 – Present
    • Senior Site Reliability Engineer Dec 2015 – Oct 2022
    • Site Reliability Engineer Sep 2012 – Dec 2015

    Systems engineer on the scalability and reliability of services including Search, Ads, News, AMP, Flights and Shopping. I work closely with Borg, Google’s job scheduling and management system, and specialize in deep-dive debugging of production systems.

    • Search ML SRE Late 2025 – Present

      Reliability and SLO compliance for Google Search serving, including the systems behind AI Overviews and AI Mode, which must stay highly available around the globe. Includes large-scale resource scheduling, capacity planning and production support on one of Google’s largest fleets.

    • Tech IRT 2020 – Present

      A member of Google’s top-tier technical incident response team: a key role in large-scale incident response across Google’s infrastructure, jumping into unfamiliar systems under pressure and coordinating across teams to minimize impact on partners and users.

    • Ads product-area resource management 2020 – 2025

      Resource tech lead for the whole Ads product area’s fleet, managing technical policy and capacity. Identified and drove Ads-wide adoption of a new caching technology that delivered durable operating-expense savings; the approach became a template later adopted by other product areas.

    • Shopping SRE 2018 – 2020

      Drove a project to migrate Shopping off dedicated machines and onto more efficient shared Borg resources, reducing its dedicated-machine footprint and realizing real savings, without disrupting a live, high-traffic product.

    • Travel Hotels SRE 2014 – 2018

      Founding member and tech lead of the team, helping stand it up from scratch and instill a strong reliability culture from day one. The Hotels product crossed a substantial revenue milestone during this period.

    • Blunderbuss 2013 – Present · 20% project

      An unfunded bug-assignment and notification system relied on by thousands of Google teams. I’ve been involved since 2013 and tech lead since about 2017, keeping it running reliably for years with no formal funding or dedicated support: a case study in high-leverage, low-overhead reliability engineering.

  2. Apr 2007 – Sep 2012

    Team Cymru, Inc.

    Systems and Software Engineer · Chicago, IL and Orlando, FL

    • Software Engineer Jul 2009 – Sep 2012
    • Senior Developer Feb 2008 – Jul 2009
    • Researcher Jul 2007 – Feb 2008
    • Consultant Apr 2007 – Jun 2007

    Valuable experience and insight into the world of information security. I developed a large-scale query tool that was an early introduction to Big Data, and had primary responsibility for building, organizing and managing a variety of bare-metal and virtualized systems, deployed both in our own datacenters and on partner premises.

  3. Nov 1998 – Aug 2006

    Dynamic Network Services, Inc.

    Founder, System Administrator, & more · Worcester, MA and Manchester, NH

    I founded DynDNS and built it into a multi-million-dollar-a-year business with well over a million customers, 100% self-funded at the time of my departure. I wore all of the hats at various times, with primary focuses on business development and overall system and network design and implementation, and was directly involved in the full systems, network and software design processes from concept to implementation.

    I even accidentally created a de facto standard: DynDNS’s “/nic/update” HTTP(S)-driven dynamic DNS record update protocol is still in use by dynamic DNS providers today, and it makes me cringe!

References available upon request.