Aumann's agreement theorem says two rational people with the same information can't agree to disagree forever. Investment committees prove this wrong every week.
I keep coming back to psychology frameworks that were never built for venture, because the behaviour they describe shows up in term sheet negotiations almost exactly as written
The founders I trust most under pressure aren't the calmest. They're the ones who can name what they're feeling while it's happening. That's interoceptive ability, and it predicts more than confidence does
I used to ask founders about their favourite user, the one who emails unprompted with ideas. Flattering question. Useless question. It tells you who loves the founder, not who needs the product.
Signups used to be the north star. Activation is now, because agentic products fail silently: a user can sign up, never get the agent configured right, and quietly walk away thinking the product doesn't work.
The real diligence question on an AI feature: is it solving a workflow, or decorating one that was already fine. Most decks answer this without meaning to.
Status quo bias is why "obviously better" products lose. Switching costs aren't rational; they're behavioural, and "better" rarely clears the bar that "familiar" already occupies.
Our daily happiness depends on:
a) Do we surround ourselves with agreeable company and more importantly, are we agreeable company to them so they like to be around us?
b) Is our mind free of suffering thoughts, which tend to trap the mind in endless loops? If I am hyper-focused on solving a problem or learning something, I am generally free of such "suffering loops" or "ego loops".
If we are trapped in suffering loops or ego loops, we tend not to be agreeable company, so (b) is a precondition for (a)
Things that make you stand out that anyone can do:
- Have good manners
- Do what you say you will do
- Put extra effort into everything, and go the extra mile.
...and make this baseline behavior, everywhere.
You will instantly build a reputation.
We're seeing an explosion of AI foundational models (open and closed-source), each with different strengths and price tags. For businesses, the real bottleneck isn't finding a model - it's deploying the right one efficiently.
That's why there is a growing appreciation for how critical inference platforms like @SimplismartHQ are right now. They are building the essential infra to make AI scalable, fast, and actually cost-effective.
VCs are quietly using AI to screen decks and do diligence before a meeting has even happened.
I've talked to three funds this month doing it. Most founders have no idea.
One of the more subtle pitfalls of starting a startup when you're too young is that you can't hire well, because you haven't had enough experience to be a good judge of people. You can judge technical ability but not character, so you end up hiring smart jerks.
Ecosystem-led growth is quietly doing what outbound used to do. If your product shows up inside someone else's workflow before your SDR ever emails, you're already winning the deal you haven't pitched yet.