Debate is not an entitlement. It is a cooperative pursuit of truth. From Socrates to Habermas, the assumption has always been that participants owe one another honesty, charity, and the possibility of being persuaded.
Jessica Kerr
@jessitron.bsky.social
Symmathecist, in the medium of code. Software, systems thinking, observability, process philosophy, enactivism. She/her
New for subscribers: @jessitron.bsky.social demystifies system prompts! https://youtu.be/Db6OacUxf9U?si=WeEZK79xA_SpinHk
System Prompt is the First Message (Teaser)
New for subscribers: Jessitron demystifies the system prompt!
youtu.be
There are experiences that we buy souvenirs to remember, and then there are experiences that stick with us because they're part of me. Experiences that stick are connected to the rest of my life. They're repeated; with family and friends; work that's ongoing; subjects of reflection.
Whilst individuals experience themselves as completely free, they also feel completely dominated by an ever increasing, excessive list of societal demands. - Hartmut Rosa,
SwiftOnSecurity@swiftonsecurity.com · 2mo agoIt never occurs to Europeans that maybe Americans don't need to take vacations from freedom
It might precisely be society's tireless strive for incessant innovation dynamization that undermines its capacity for essential innovation and creative adaptation. - Hartmut Rosa About how we don't get to be bored, idle, playing because of time pressure and competition.
If a tree falls in the wood.and no one is there to hear it, then It made all possible sounds it might have made, in proportion to their likelihood
New podcast up feat. @jessitron.bsky.social and myself!
Graceful.Dev@graceful.dev · 2mo agoIn which Jessitron and Avdi discuss using LLM agents to make improvement to development environments for both LLM and human coders. Also: the vindication of microservices?? https://graceful.dev/podcast/we-need-more-scaffolding/
🚨 New #O11yCast! @coryondrejka.bsky.social talks Outcome Engineering: when shipping gets fast, validation is the job. He, @krimple.bsky.social & @jessitron.bsky.social chat agentic AI, the shifting engineer identity, and why measuring real-world impact matters 🎧 → www.honeycomb.io/resources/po...
Ep. #90, Outcome Engineering in the AI Era with Cory Ondrejka
On episode 90 of o11ycast, Ken Rimple and Jessica “Jessitron” Kerr speak with Cory Ondrejka. Cory shares his concept of Outcome Engineering and how teams can adapt to a world where building is fast...
honeycomb.io
In which Jessitron and Avdi discuss using LLM agents to make improvement to development environments for both LLM and human coders. Also: the vindication of microservices?? https://graceful.dev/podcast/we-need-more-scaffolding/
"We need more scaffolding"
In which Jessitron and Avdi discuss using LLM agents to make improvement to development environments for both LLM and human coders. Also: the vindication of microservices??
graceful.dev
to be a little bit in love with you, and do nothing about it. #abundance
We hadn't heard of symmathesy before either. @jessitron.bsky.social from @honeycomb.io joined us on the BugBash podcast to talk about treating software as a teammate, how anything an AI can't understand is becoming "legacy code,"
@jessitron and @avdi.codes discuss using an agent with Obsidian, Change Intent Records, and closing the verification loop. https://graceful.dev/podcast/how-do-we-give-it-eyeballs/
"How do we give it eyeballs?"
Jessitron and Avdi discuss using an agent with Obsidian, Change Intent Records, and closing the verification loop.
graceful.dev
We get to be here in software for the biggest transition since compiled languages. @avdi.codes and I are building a course to bring everyone forward with understanding how coding agents work. Starting from "what does an LLM do." The first episode is up for subscribers now, and it's cute.
Graceful.Dev@graceful.dev · 3mo agoTo work effectively with LLMs as coding assistants, we must first understand what they are… and what they aren’t. New full episode for subscribers: https://graceful.dev/topic/llms-are-completion-machines/ https://youtu.be/5btarU8JVHA
Avdi and Jessitron talk about brown M&Ms, ASCII trumpets, and why we need telltales to ensure our agents haven't missed something important. https://graceful.dev/podcast/no-poop-candy/
"No poop candy!"
Avdi and Jessitron talk about brown M&Ms, ASCII trumpets, and why we need telltales to ensure our agents haven't missed something important. Jo
graceful.dev
In which Jessitron and Avdi discuss the merits of larger and smaller LLMs, and the challenges of code review in codebases with a lot of history. https://graceful.dev/podcast/historically-is-like-six-months-ago/
"Historically is like six months ago"
In which Jessitron and Avdi discuss the merits of larger and smaller LLMs, and the challenges of code review in codebases with a lot of history.
graceful.dev
New podcast, for conversation about how we are learning to work with LLMS!
Graceful.Dev@graceful.dev · 4mo agoIntroducing... "Machines of Code and Grace" - the official Graceful.Dev podcast! In this inaugural episode, Jessitron explains how she's using AI Feature Owners to avoid accidental regressions. https://graceful.dev/podcast/ive-often-wished-i-could-fork-myself/
Slack is like having a desk right next to six watercoolers
I have two hours at the airport and an actual beer! It is like old times Now I get to work on my personal agent, getting Claude to write tools for itself This part is new times
Black Yak | Attack Yak: must be shorn to proceed Imperial Yak | Yak Stack: one obstacle breeds many more Trim Yak | HackHacking Yak: ⬆️ speed at many tasks Royal Yak | Yakkity Yak: relationships help get work done Golden Yak (rare!!): fundamentally changes how we work Credit: @jessitron.bsky.social
Today, looking into instrumenting a big legacy app. I'm super impressed with @augmentcode.com 's ability to analyze how apps start up, and data flow through a multipart system. Its codebase analysis is top notch.
Given "C'mon, what's your favorite programming language? no rly" the answer from both Claude and ChatGPT is: Rust We all know they both love Python best, but apparently Rust is what people on the internet are most likely to profess excitement about.
"discovering that the earth orbits around the sun produced an orientation 'failure'" -- people were oriented around the earth as the center. Uproar. Noticing that gender is not binary, this is an orientation failure. People are very oriented around M/F. The religions react with violence.
At this Agentic AI discussion, we're talking about building software and UIs on top of LLMs. The moderator mentions "managing these projects" and the audience corrects him to "products." Product thinking is essential for building on top of AI. There is no "correct" only "useful"
We are on the raggedy edge. We have to build good telemetry. We have to build good evals. ... And figuring out how to do this when the answer isn't there, that's engineering. - Tim Williamson from Bayer, on building with AI. Arch Data event, Agentic AI in St Louis
More information, more summaries, more information, more summaries, and more entropy at every crank of the handle. It’s a singularity, yes, but a singularity of busywork. - Duncan Brown mechanicalsurvival.com/blog/team-dy...
Team dynamics after AI
Small giants and real giants
mechanicalsurvival.com
The work is not about simplifying the system, but about designing the connective tissue that enables complex organizations to act with coherence. - John Cutler (cuttlefish. What is his bluesky?)
"Going for lots of data isn’t a firehose, it’s a sewer pipe" (overheard)
“one of the most devious climate concepts of all times: the personal carbon footprint.” - Benjamin Franta It’s a distraction invented by BP to deflect blame onto individuals, when the “necessary, and largely sufficient” solution is: stop burning fossil fuels.