Miroslav Lysyuk

Product leader & sensemaker with a bent for anthro & philosophy. Decade+ launching in IoT @samsara, AR/VR @mapbox, energy, & more. Hardware art for fun.

[Miroslav Lysyuk @miromusing]

I've been interviewing a variety of consultants about their use of LLMs in their workflows—and the insights are eye-opening! From product to marketing to sales enablement, the way AI is integrated varies wildly.

[Miroslav Lysyuk @miromusing]

@SenWarren Instead of celebrating the people that are a reflection of what makes our country great, you’re wondering how to villify and undercut them. It’s really disappointing.

Meanwhile, how’d you make your millions? How many people did you employ? What new industries did you foster?

[Miroslav Lysyuk @miromusing]

@simonsbookclub Well said, and the irony is vilifying those that earned it because they didn’t use the money to [insert self reflexive money shot daydream].

[Miroslav Lysyuk @miromusing]

@bourscheid I can’t tell what’s more annoying, the false equivalencies that have nothing to do with one another, or your repetitive whiny posts about Elon to engagement farm.

[Miroslav Lysyuk @miromusing]

@mrhawking123 @HowToAI_ When this paper came out, I was super excited, and built https://t.co/6BjPY18Sx5 to apply the method. You can see what I mean by playing around with it.

[Miroslav Lysyuk @miromusing]

@mrhawking123 @HowToAI_ LLMs give statistical average responses to whatever framing you pose. I think that's definitely useful for getting quick average signals on any idea you have. The risk is that it ends up being shallow, predictable, and lacks the nuance and freshness that real feedback gets you.

[Miroslav Lysyuk @miromusing]

@greymerchant00 @HowToAI_ Yup - I tried to replicate the method and didn’t see the same results. You can try it at https://t.co/6BjPY18Sx5

[Miroslav Lysyuk @miromusing]

Hype is a funny thing. Like watching religious rituals masked in product reveals: the conversion testimony ("I tried it and everything changed"), the coming salvation at takeoff, the heretics rebranded as doomers, the founders talked about the way saints used to be.

[Miroslav Lysyuk @miromusing]

@gabepereyra Not needing to rely on the foundation model providers cuts token costs and reduces data & IP loss concerns.

I agree with your broader point about companies realizing they need to build proprietary tooling - especially in services. We're doing that at our firm right now.

[Miroslav Lysyuk @miromusing]

Finally being a weird mix of designer, product manager, and developer is having its moment with Builders.

[Garry Tan @garrytan]

There is a transition here for people across the workforce: the working world needs fewer measurers and more builders

More revenue means there will be more activity and more building, and in the shorter term less measuring

[Chamath Palihapitiya @chamath]

Agreed!

This is what Software Factory helps enable - it organizes and creates the artifacts that guide all downstream productivity.

Increasingly we are finding that large enterprises want a control plane that organizes the tribal knowledge while also providing governance, safety and control.

This seems to be accelerating as the models asymptote on quality.

Try it here: https://t.co/fkfTXgdfXK

[Miroslav Lysyuk @miromusing]

@imidaily B

[Miroslav Lysyuk @miromusing]

Both Jaguar and Ferrari have recently made the same mistake - their product design process got disconnected from their brand which got disconnected from their market. Classic operational product market fit drift and a sign of internal decay in the face of market shifts.

[Aaron Boodman @aboodman]

This is why the reaction to the Luce is so negative. The styling is beside the point. The interior is beside the point.

The problem is that Ferrari hasn’t made the case that this car preserves the core heritage behind the prancing horse – a badge with *decades* of love built up.

[Miroslav Lysyuk @miromusing]

One of the most under-appreciated and delightful types of vibe coding is with hardware.

For instance, here, in 5 minutes I spun up a real time map of global lightning strikes as art 🖼️ 😍🦾

Get a cheap LCD + WiFi board and go crazy.

[Miroslav Lysyuk @miromusing]

Yes but it’s worth noting that this argument is based on the assumption that the cost per token stays fixed and the tokens per task stay fixed. Both are not necessarily true, and if we want to use industrialization as an example, the cost of machines went down over time and efficiency went up.

[Miroslav Lysyuk @miromusing]

@signulll The best strategic response is to leverage the existing logos & trust you have (against bottom of market) and build a niche-specific data & context moat that the model companies can't (yet).

[signüll @signulll]

every software incumbent is fighting a dual front war right now, & dual front wars are historically challenging af.

from above, model companies are vertically integrating through their own surfaces... you can see anthropic, openai, google are all shipping products that directly compete with established saas categories. from below, an effectively infinite number of small teams are filling niches faster & cheaper than ever before, cuz the minimum viable team to build real software has collapsed to 1-3 ppl with ai.

the squeeze this creates is existential in so many ways & you can see markets are pricing this squeeze in now. incumbents can't focus resources upward without getting hollowed out from underneath, & they can't play whack a mole with disposable startups without leaving themselves exposed to platform level moves from above. the standard playbook of acquiring threats, raising switching costs, & bundling feature doesn't really work when the bottom up competitors are infinite & don't need to capture much value to survive.

this is the fundamental structural shift most people haven't internalized about what's actually happening with ai. quite simply ai is simultaneously compressing the entire competitive landscape from both ends, & the companies caught in the middle have no really good strategic response.

[Miroslav Lysyuk @miromusing]

Larger social groups diffuse knowledge and technology faster, even when you aren’t close to the other parties.

[Miroslav Lysyuk @miromusing]

A fascinating aspect of neighborhood WhatsApp groups is that the moment they begin to exceed 100-150 or so people, they inevitably shift from connection to commerce.

[Miroslav Lysyuk @miromusing]

This even when the moderators try to keep the original spirit alive by discouraging commerce.

There’s a very human principle at play here.

One theory for why Homo sapiens were a successful species relative to other hominids like Neanderthals is our social groups were larger.