In previous posts, I've described LLMs as "helpful liars." This is useful mental model, but it doesn't really help answer the question, "how do you get them to lie less?" After using LLMs 15+ hours a day (plus a bunch of my own independent research on exactly this) I've developed a new mental model...
I've been exploring OpenClaw, but giving so much access to an autonomous AI agent with 50+ connectors, persistent memory, arbitrary code execution, and 512 reported vulnerabilities that can impersonate me by design has made me uneasy. So I wanted to explore a safer version. What "safer" means to me...
This post is the story behind the research. If you want the full paper with methodology, statistics, and raw data: Receipt-Gated Pipelines on GitHub. I caught three AI models fabricating security reports. Complete with CVE numbers. Severity ratings. Remediation advice. For vulnerabilities they...
This post is the story behind the research. If you want the full paper with methodology, statistics, and raw data: Structure Beats Scale on GitHub. This shouldn't have worked. I took a model that costs a tenth of a cent per call — Mercury 2, a diffusion-based reasoning model that nobody was talking...
Seven days ago I wrote that the main thing that reliably matters in AI-assisted development is quality gates. Nineteen sprints across three projects, five falsification experiments, and the punchline was "just run your damn tests." But I left out the harder question. If implementation velocity is...
Last time I wrote about Flowstate, I was ten sprints in across two projects. I had a workflow that worked, hypotheses that seemed right, and the conviction that skills, waves, and multi-agent delegation were the key ingredients. I was wrong about most of that. Nine more sprints, a third project,...
I built an encrypted task manager with AI agent teams. I wrote about it. People seemed interested in the workflow — the skills, the waves, the retrospectives. What I didn’t tell you was that the workflow was held together with duct tape and copy-pasta. Every sprint, I was the bottleneck. I designed...
I’ve been a Product Manager for 20 years. Design Within Reach, MyFitnessPal, Tonal, Habitry, a cybersecurity startup that Mimecast acquired, a VR fitness app that Meta acquired. On paper, wildly successful. In practice, I’ve been pretty bored for the last six years. Not unhappy. Not ungrateful....
When I think, I write. Most often with a pen, staining a piece of paper with ink. I have amassed a collection of pens, inks, and paper with which I do my best thinking, and people are surprised to hear that I use different combinations of these tools to think in different ways. The bulk of my...
I am Jeanne Varnell’s eldest grandchild. We called her “Boo.” A perfect, unique grandparent name that makes everyone smile as much as she made us smile. If you know anything about Jeanne Varnell, it’s that she was generous. She — like the generations of Hyde women before her — changed the lives of...
Recently, Habitry was approached to design a mental-health chatbot. The company is quite new and didn’t have a pre-existing design process that we were slotting into, so they basically asked us, “how do we even Motivation Science?” from scratch. This was a fun problem to think about and we thought...
“The most ordinary conversations are fraught with life and all its meanings.” — John Haugeland, Having Thought Habitry gets asked to help out with a lot of chatbot design. We got our start teaching health coaches about how to have need-supportive conversations with their patients and clients, so...
This weekend, two giants in the field of Behavior Change Research, professors Martin Hagger and Mike Weed had a debate at the 2018 Meeting of the International Society of Behavioral Nutrition and Physical Activity in Hong Kong. (Oh, you didn’t know? Did you miss the Super Bowl commercial?) The...
On Thursday, May 3rd 2018, my wife and I got a call from her Father that started with, “where are you?” He was calling to tell her that he and her brother were OK, but at 1:00 in the morning the home my wife grew up in had burned to the ground. The home Leon had spent more than 40 years in was...
In 1995, I became a man. I don’t mean that in a creepy way (I was 12), but in my 12 year-old mind I became a man because that Summer while at Skymont Boy Scout Camp, I passed the Knife Safety Course. And it was this achievement of getting all the answers on a multiple-choice test which granted me...
I read a few non-fiction books a week. At any given time, I’m actively reading between 3–10 of them. And since it’s something I’ve been doing since 5th grade, I thought I’d about write how I do it. I’ll get into logistics, but first let’s talk about goals. In grade school, there was a program in my...
The Greek word for philosopher (philosophos) connotes a distinction from sophos. It signifies the lover of wisdom (knowledge) as distinguished from him who considers himself wise in the possession of knowledge. This meaning of the word still endures: the essence of philosophy is not the possession...
The other day on Facebook, Mark Schneider asked a great question: <blockquote "Within the SDT/OIT concepts - what defines a “group” or “community”? Is it as loose as “what the person FEELS/perceives” to be part of? Or is there a more concrete idea that may have been used for the research on this...
Habitry is officially partnering with Lift the Bar! Get ready for the biggest hands-on autonomy support program in the world. For those of you who aren’t familiar with Lift the Bar, it’s a place for coaches to get daily coaching, education and support for Personal Trainers. We have been working...
Good news for everyone who loves our podcast, the bettercast, and the work Habitry has been doing for over two years! The short version: As of today, Betterish is here! We’re doubling down on exploring real stories of change. You might’ve noticed we stepped up our podcasting to one episode a week...
Stevo's Blog
Writing on product, behavior change, and various personal musings.