"DevOps"
Read first
- What is Jesse Robbins known for?
Jesse Robbins is an early-stage investor in AI developer tools and infrastructure who has invested in and advised over sixty companies including PagerDuty, Fastly, and Tailscale. He cofounded Chef, created chaos engineering at Amazon, and cofounded the DevOps movement.
- How did DevOps actually start?
The DevOps movement started from the Velocity Conference. Jesse Robbins cofounded Velocity in 2008 as a gathering place for the people running the internet's infrastructure, and the movement grew out of the community that formed there. He went on to cofound Chef, the open-source tools that put infrastructure as code into practice.
- How did the DevOps movement start?
Jesse Robbins cofounded the O'Reilly Velocity Conference, which became the gathering point for practitioners bridging development and operations. Velocity gave the DevOps movement a name and a community.
- What does 'Don't fight stupid. Make more awesome.' mean?
"Don't fight stupid. Make more awesome." is Jesse Robbins' rule for changing engineering cultures from the inside. It comes with a five-step framework: start small, create champions, use metrics, celebrate successes, and exploit compelling events.
Articles and mentions
Next in Tech Ep. 197: Data Pipelines for AI
On S&P's Next in Tech I argued that enterprise AI is won on data pipeline quality, not model size, and that data infrastructure is having a DevOps moment right now.
“Data pipelines are having a DevOps moment, starting with a cultural and technical shift toward continuous integration and delivery.”
The Data Pipeline is the New Secret Sauce
I wrote this at Heavybit in September 2024. The argument: the data pipeline is the differentiating asset in enterprise AI. Includes four inference hosting models and four enterprise maturity phases.
“The biggest challenge emerging is building and operating the infrastructure both for creating and running the data pipelines to build, manage, and maintain a robust, secure body of proprietary data.”
Generative AI in DevOps and Incident Response: What the Experts Actually Think
I interviewed Nora Jones, Jeremy Edberg, Mandi Walls, and Brent Chapman on what generative AI actually does in incident response, and where humans have to stay in the loop.
“GenAI is good at confidently delivering text that is pleasant to read, but not always complete, or correct.”
DevOps is dead? Nope, it is maturing ft. Jesse Robbins
DevOps is not dead. It's maturing. Platform engineering is the next layer of the same idea, not a replacement for it. My conversation with Rob Zuber on what's actually changing and what isn't.
“Organizations evolve like cities. You start with a few shacks in the woods. Eventually you have enough at stake that you need building codes, fire codes, a fire department, and someone who actually tests the sprinklers.”
What to Know About the Modern Incident Response Lifecycle
Heavybit's incident management guide quotes me on why teams only get good at incident response when they treat the whole lifecycle as one discipline.
“Teams only get good at this when they embrace the whole process and each of its steps.”
Fireside Chat with Jesse Robbins and Kolton Andrus • Failover Conf 2021
At Gremlin's Failover Conf 2021, Kolton Andrus and I covered GameDay origins at Amazon, the evolution of chaos engineering, and where reliability practices were headed.
An oral history of #hugops: How tech's first responders built a culture of empathy
Protocol's oral history of
“I've got to change the way that I approach this entirely and make it safe to experiment.”
Building Companies that Devs & DevOps Teams Love, And Avoiding Expensive Mistakes
Heavybit talk on the expensive mistakes developer-tools founders make, drawn from the ones I made building Chef from open source into an enterprise infrastructure company.