Profile avatar
ferd.ca
Staff SRE @ honeycomb.io, Tech Book Author, Resilience in Software Foundation board member, Erlang Ecosystem Foundation co-founder, Resilience Engineering fan. SRE-not-sorry. blog: https://ferd.ca notes: https://ferd.ca/notes/
169 posts 1,404 followers 156 following
Regular Contributor
Active Commenter
Posts 20 Comments 30

I realized something recently, which is that we're building AI tools completely backwards. While I was aware that there's some de-skilling happening already, I finally clicked a bunch of pieces together and decided to write this to talk about how we can do *better* hazelweakly.me/blog/stop-bu...

submitted 2 days ago • 24 comments

This is also why I push for engineering organizations to have EVIDENCE strategies not just "measurement" or "metric" strategies. Evidence strategies include qualitative insight! Different FORMS of evidence and understanding whether we are or aren't getting them is evidence strategy!

submitted 3 days ago • 2 comments

corgi is the plural of corgus

submitted 4 days ago • 1 comment

Gonna have to create a new kind of Quebec-centric sci-fi dystopian genre focused on AI, inequity, and societal collapse, and it will be called “ciboire punk”

submitted 19 days ago • 4 comments

The best customers are ones that make your product better. Huge thanks to the @honeycomb.io team for co-designing new PagerDuty insights with us! Special shout-outs to @ferd.ca and @another-engineer.bsky.social in particular. 🙏

submitted 25 days ago • 1 comment

SRECon Americas videos are up, which means you can watch the closing plenary @charity.wtf and I gave—an open letter to AIOps vendors if they are to write better tools. It also turns into an open letter to SREs, because we don't expect vendors to all listen to us. www.usenix.org/conference/s...

submitted 26 days ago • 0 comments

I was trying to write a quick post here about how I'm taking more time off my usual online stuff after how demanding the last few months were, only to find myself writing more and more and going "this could be a good work blog post" and oh yeah of course I'm in this situation, I just keep doing this

submitted 31 days ago • 1 comment

overhearing LLM programming chat and this is all I can think about

submitted 40 days ago • 0 comments

While people can feel like they’re getting better and more productive with tool assistance, it doesn’t necessarily follow that they are learning or improving. @ferd.ca www.honeycomb.io/blog/ai-wher...

submitted 42 days ago • 0 comments

I know all this power and identity stuff is hard. But don't you just want to TALK talk about what it means to be in tech right now? What it really all means and why this is happening to us? Because I do. We need the talk story. It is the only way we can find something different

submitted 42 days ago • 2 comments

I really am growing tired of the "be careful" words of warning from the folks running the knives-without-handles factory.

submitted 49 days ago • 0 comments

Who's coming to SRECon this week?? @ferd.ca and I are doing the closing keynote! 👯 And @honeycomb.io is hosting a happy hour on Wednesday along with our friends from @buildkite.bsky.social, @spacelift.io, Oso and Cloudsmith. Register here 👉 events.spacelift.io/off-call-on-... ... see you soon!!

submitted 55 days ago • 1 comment

This week's paper, digging further into Work-as-Imagined and Work-as-Done through Lucie Cuvelier's text, comparing them in what I loosely translate as "when resilience engineering questions ergonomics once again," which compares Resilience Engineering's perspective with French Ergonomics concepts.

submitted 56 days ago • 1 comment

People tend to have a mental model where a system is stable until disturbed, far more often than they have one where the system is balanced because it is constantly intervened with. The latter is a more useful approach to thinking about complex systems.

submitted 63 days ago • 1 comment

Despite all the stereotypes, sometimes software gets the math wrong 💀 Why care about understanding how software metrics operate in the real world at scale? Because I see tech folks constantly talk big about using terrible aggregations that make no sense. arxiv.org/abs/2503.05040

submitted 66 days ago • 1 comment

another version of this: sitting down to write: I don’t know what to say, I don’t have anything useful to say getting asked a question in the dms: here’s a war and piece length answer to your question and every follow up to every followup I could imagine, delivered in 4-400 word chunks

submitted 68 days ago • 0 comments

Any random tech discussion: this off-hand comment someone made is a perfect opportunity for a very opinionated rant that could run for hours Preparing a tech talk: it will take many weeks to properly structure ideas into a coherent narrative worth sharing with others

submitted 68 days ago • 2 comments

This is the stuff that you find in practice when you try to do something that's not closely tied to an industry standard benchmark. And also, a dire warning sign for any business who does find that they get a much higher success rate from OOTB models.

submitted 71 days ago • 0 comments

I recently read the paper "Towards Joint Activity Design Heuristics: Essentials for Human-Machine Teaming" which I loved so much I wanted to make it easier to share. To that end, I've excerpted the Ten Heuristics from the paper here: human-machine.team with anchors for each heuristic.

submitted 73 days ago • 0 comments

You heard of AIOps, and of having LLM agents roaming into your infrastructure? I've written a post with important questions you should ask about their design and how they interact with your system: Where in the Loop Should Humans Go?

submitted 77 days ago • 0 comments