From the desk of Sean Heilweil
Operator Note
I did something this week I hate admitting.
I was the blocker. The bottleneck. The guy with a list of reasons why something was not ready yet.
My founder in residence Emi vibe coded a new website for Sur. She wanted to ship it. I spent two weeks stopping her. Meta descriptions. Sitemaps. Analytics. llms.txt. I had a list of 67 things that "needed" to happen first.
You know what that list actually was? Fear wearing a project management costume.
On our weekly call I finally snapped. "Fine, do whatever you want." That is when Emi clapped back: "No!! We do not work this way together!"
She was absolutely right to check me.
Here is what a couple decades of building actually taught me that I forgot.
Pattern recognition is the most dangerous skill a senior operator has. Not because it is wrong. Because it is right just often enough to justify killing someone else's momentum every single time.
I have launched thousands of sites. I know what breaks. I know what gets missed. I know what is going to go wrong before it happens. And none of that matters if the person you are working with never gets to learn it themselves.
A lot of experienced founders think mentoring means preventing mistakes. It does not. It means creating the conditions for someone to develop their own pattern recognition, which requires letting them build the scar tissue.
The founder in residence model only works if you let people fail. Not "fail safely with guardrails and a 67-point checklist." Actually fail. Ship something imperfect. Watch what happens. Learn from the texture of real consequences.
So I backed off. Emi shipped it.
And I spent the next few hours clicking through every page, reading every line of copy, getting genuinely excited. Not because it was perfect. Because it was better than what came before it. That is the only bar that matters on iteration one.
That is the part most senior operators miss. They confuse quality control with control. They dress up fear as diligence. They call their bottleneck behavior "mentoring" and never notice that the person across from them has stopped shipping.
The real damage is not a website that launches without perfect on-page SEO. The real damage is a talented person who stops trying because someone with "experience" made them afraid to move.
There are exceptions. Of course there are. Some things genuinely need a checklist. Compliance. Security. Anything where failure creates irreversible harm. But a marketing website on iteration one is not that. And most of the things experienced operators block are not that either.
If your expertise is preventing someone else from getting theirs, you are not a mentor. You are a bottleneck with a title.
What I'm Seeing
Senior operators confusing oversight with ownership. You can review work without holding it hostage. A lot of founders have not figured out the difference.
Vibe coding is lowering the cost of iteration to near zero. Which means the penalty for shipping something imperfect has never been smaller, and the penalty for not shipping has never been higher.
The best founder-mentor relationships have explicit permission structures. Not "I trust you" as a vague sentiment. Actual agreements about who has the final call on what.
A pattern I keep seeing in residence programs: founders build incredible things in week one, then slow to a crawl by week four. Usually because someone senior started "helping."
Perfectionism in early-stage companies almost always traces back to a single person. Find that person. That is your bottleneck.
Behind The Scenes
I have been thinking a lot lately about the difference between earned intuition and borrowed fear.
When I blocked Emily, I told myself I was protecting the work. But I was actually protecting my own comfort. If the site launched with issues I could have flagged, what would that say about me? That I did not do my job? That I let something ship that I knew could be better?
That is ego dressed as diligence. And it is one of the hardest things to catch in yourself because it genuinely feels like caring.
The moment Emily pushed back was the moment I realized I had been optimizing for my own reputation, not her growth. That is an uncomfortable thing to sit with. I am still sitting with it.
Tactical Idea
Run a Bottleneck Audit this week.
Look at every project or deliverable that is currently waiting on your approval, your review, or your input. Write them down. Then ask three questions about each one:
What is the actual worst-case outcome if this ships without my involvement?
Is that worst case reversible?
Am I blocking this because it genuinely needs me, or because I am uncomfortable not being involved?
If the worst case is reversible and you are blocking it out of discomfort rather than necessity, let it go. Send a message today. "Ship it. I trust your judgment. I will review after it is live."
The question worth sitting with: How many of the things on your plate are there because they need you, and how many are there because you need them?
Closing Thought
Experience is supposed to be a gift you pass forward, not a gate you stand behind. The moment your pattern recognition becomes someone else's paralysis, it has stopped being useful.
The bottom line: Mentor by creating space, not by filling it.
— Sean, Cache CEO