Writing
I picked up my laptop by the corner of the screen last week. Turns out that's not the right way to move a laptop. The screen cracked.
I've been to several startup pitches a week for the last couple of months. It’s been a while since I’ve operated in this space regularly, but man, the conversation has not changed at all.
Companion piece to 'The VC model is broken for founders.
Direct observation from a conversation with a founder at zero customers and zero revenue asking for VC introductions.
Structural misalignment between fund incentives and founder incentives. Opens with power-law math, lands on the AI-native alternative that removes the need for the swing.
Georgia SB 406 passed the state legislature on March 31, 2026. Senate vote: 51 to 0. House vote: 155 to 10.
American philanthropy moves roughly $500 billion per year. That number looks like generosity at scale. What it obscures is how much of that money never reaches the people it is meant to serve.
A development rule about surgical changes over wholesale rewrites, and what it reveals about the difference between institutional memory and organizational folklore.
The deployment for a new product was built on Cloudflare Workers.
We were walking along the river near our house earlier this week watching ducks on the water. Ashley mentioned a video she had seen of ducks swimming from an underwater camera.
Operating rules encode assumptions. When the assumptions change, the rules expire. Most founders never notice.
Most companies adding AI to their products are becoming more efficient. A different category of company is only possible because AI exists at all.
Founders are the worst proxy for their own customers, and this is especially dangerous because they are smart, motivated, and convinced they understand the space.
Five tools. One job each. No overlap. A brief case study of how the Beacon agents runs cold email for Rebuilt. Apollo finds gym owners by business type, title, and company size.
Most operating systems fail the moment the company they were built for stops looking like the company using them.
There is a category of business relationship that most founders have tried and most founders have found disappointing. The advisor relationship.
Solution attachment is natural, emotionally satisfying, and professionally dangerous. The problem-first orientation is a discipline, not a personality type.
The first version of every product I've built runs on Railway. Here's why: Railway does one thing well: it takes a GitHub repo and runs it. You push code, it deploys.
We've all seen Company Values. They are printed on the office walls, embedded in the onboarding deck, referenced in the all-hands meeting once a quarter.
A friend and I had lunch recently. I was telling him about all the conversations I have with people looking to build, but struggling to find product market fit.
Most leadership development programs are designed to develop people into the leader the organization wants.
New markets are won by relationships, not programs. This is true whether the market is a new geography, a new customer segment, a new community, or a new industry.
The moment in a fundraising conversation is not when you ask. It's everything that happens before it.
The first 30 days in any new role have one job: observation. Not change. Not fixing. Not proving yourself. Watching.
There is a version of organizational failure that looks exactly like success from the outside.
Every organization has a version of this problem, or it’s inverse:
A few years ago, if you had a product idea and no technical co-founder, you had a problem. You needed someone who could build, or you needed money to pay someone who could. That was the gate.
Allyship that flows in only one direction is not allyship. It's charity. And charity, over time, breeds resentment.
Most organizations are using AI as a drafting assistant or documentarian. Those are among the least interesting things it can do.
The Army runs on a simple hierarchy: mission defines what you're working toward. People are how you get there.
The word "business" comes from the Old English bisignes — the state of being busy. Occupied. Engaged in activity. That is the root of the thing we have built entire economies around.
There's an old distinction in performance: the amateur practices until they get it right. The professional practices until they can't get it wrong.
I have a complicated relationship with running. The Army deserves most of the blame for that, having used it as both training and punishment, sometimes simultaneously, and the association stuck.
Slow builds are expensive. Expensive builds die. Founders run out of energy. Companies run out of runway. Significant others run out of patience.
It is a poor craftsman who blames their tools. But also, shitty tools suck. Emmett, the venture engine that runs the operational side of Meridian, is named after my great-grandfather.
If you want to see it: [ForgeAI Demo] If what you typed made you think "I should actually build that": [email protected]
Creative Shit project. Language spec at codethejewels.com. Repo: everettsteele/code-the-jewels
Ask almost any leadership team what their priorities are and you will get a list.
For a long time I used AI assistants the way most people do: open a chat, ask a question, get an answer, close the tab. This was cool, and it provided value.
The pitch for Kanga was simple. You're at IKEA. You just bought bunk beds. They won't fit in your car.
Meridian is a venture studio building AI-native products. neverstill.llc
The most dangerous place to start a business is right after a win. Not because success makes you reckless, exactly. It's more subtle than that.
When I receive a meeting invitation without an agenda, I decline it. This is not a quirk or a policy — it is a direct expression of how I think about work.
Most businesses aren't built to harm people. That's not the problem.
My Raspberry Pi 5 cost about $80 on Amazon. I bought it to block ads on my home network. It didn’t do a great job at that. Then I turned it into a remote controller for a 3D printer.
Four hours later, SORTED was live.
A basic income floor stabilizes demand. People spend money when they have it. Businesses depend on that.
At Meridian, every instruction set, every workflow, every architecture decision lives in Notion. Not as documentation after the fact. As the source of truth before anything gets built.
In my prior role, I often sat in meetings where someone was walking through a problem and I would simply ask: why? They'd explain. I'd ask why again. They'd go deeper. I'd ask again.
Every organization has its own version. Processes nobody designed on purpose. Approval chains that made sense once, in a meeting three years ago, for a reason that has since evaporated.
I learned a lot of things in the Army that I use even now, 20+ years after I joined. I picked up lots of cool techniques, and skills, and creative profanities.
The default ambition in startup culture is to raise money, grow fast, and figure out the business model later. That's a valid path for some founders building some things.
As a musician, I learned early that people will play whatever sheet of music you put in front of them.
Most of the frameworks people use to organize their work life treat it as a ladder you climb. Each rung is a bigger title, a bigger check, a bigger company.
