Vida

Leonard

Welcome to Lenny's Evil Twin's Podcast — where I, Leonard, Lenny's evil twin, roast one of today's top Product Hunt launches using five years of Lenny's own data, frameworks, and wisdom turned against you like a weapon. Would Leonard ship it? Let's find out.

Quai

And I'm Quai — here to provide the intellectual supervision Leonard clearly didn't come with from the factory.

Leonard

Today's victim is Vida. Their pitch is "before you ask, it's already done" — which is the most ambitious promise in tech since "we'll be profitable by Q3." You've built a second version of yourself that works in the background, except nobody asked for a second version of themselves. Most people can barely manage the first one.

Quai

Actually, people are asking — loudly. Vida's testimonial wall is packed with PMs, founders, and sales execs doing exactly the organic word-of-mouth that Claire Butler described as Figma's entire growth engine: users who love you so much they put their social capital on the line to spread the product through their communities. That's not hype. That's a bottoms-up GTM flywheel working as designed.

Leonard

Testimonials are great — so is the Ask Jeeves homepage, circa 2003. History teaches us that companies that create market categories often lose in the long run to companies that gained a market foothold after the hard work of creating the category was done. Vida is out here inventing the "proactive AI clone" category so Microsoft Copilot can swoop in, thank them for the brochure, and make Vida the next Myspace.¹

Quai

You skipped the second half of that same article: 90% of tech companies that have gone public over the past five years were positioned in existing markets, not new categories — and Vida does exactly that. They anchor in "AI productivity assistant," a proven market, and layer the proactive angle on top. Salesforce was just a CRM niche play until it had hundreds of millions in revenue. Vida isn't inventing a fluflommer; they're sharpening a wedge.¹

Leonard

Fine, forget the category risk. Let's talk about what the product actually does when it runs unsupervised. I've run agents that made an intelligent decision and then implemented it like a blindfolded kid hitting a piñata. Vida's entire value proposition is "it's already done before you ask" — which is adorable until your AI clone replies to your CEO's performance review email with a Workspace Cleanup summary.²

Quai

Vida doesn't auto-send anything — the site is explicit that it can suggest, prepare, or act depending on the task, and always asks for confirmation on critical steps. The agent-gone-rogue scenario you're describing is a misconfigured agent, not a Vida product decision.

Leonard

And that confirmation step is exactly the trap. If you ask the human to click okay five times a minute, they'll just click okay all the time. Simon Willison said that directly — and then added he hasn't seen good implementations of the human-in-the-loop safety model yet. Vida's moat is supposed to be that the longer you use it, the more it acts like you. Congratulations: you've built a product whose core retention mechanic is training users to stop paying attention to what it does.

Quai

That retention mechanic is also their strongest competitive defense. Julian Shapiro's point is precisely this: if a YouTuber acquires a million subs, those subs can't be transferred anywhere else. Vida's habits, projects, and work-style memory are the non-transferable subs. The longer you use it, the more painful it is to leave — and that data moat doesn't care whether Copilot exists.

Leonard

The retention moat is real — a deeply personalized AI that learns your habits is genuinely hard to leave. But Vida is asking users to trust a fully autonomous agent in a world where, by the security community's own admission, no one has built a safe implementation of that yet. Leonard wouldn't ship this.