Daniel Filler

Big news: I’ve joined Affirm as Director of Product Design (D2C)! 🎉

Money’s messy—but it doesn’t have to be confusing. I’m excited to keep building honest tools with a team on a mission to make finance more human.

Let’s do some good. 💙

Grids, cool. Progressive Blur, yass!

As a Framer customer, Figma Sites should be a breeze to pick up. As a Framer customer, I feel for Framer.

Figma Sites feature request:
Import from Framer

Designers don’t want handoff.

Devs don’t want static files.

The line between them? Already blurring.

Figma’s real opportunity isn’t better mockups — it’s making what we design real faster than ever.

Looking forward to today.

Monopoly, breaking up families for 90 years.

I stand with Ukraine. 🇺🇦

You can invest in good design now—or pay for bad design later.

Case in point: Sonos lost $900M in market cap after the botched launch of its new app. A brand synonymous with great UX became a cautionary tale overnight.

The ROI of design can be hard to quantify, but the cost of bad design is crystal clear.

Cutting design in the name of efficiency might feel right in the moment, but companies that over-index on this will pay for it later—because less design often means bad design.

“The question of whether design is necessary or affordable is beside the point. Design is inevitable. The alternative to good design is bad design, not no design at all.” — Doug Martin

The opposite of user-centered theatre?

I don’t buy the argument that AI is tech in search of a problem. Innovation is about adding value in novel ways. LLMs unlock new capabilities—enabling us to solve real problems in ways we couldn’t before.

Sure, there will be crap. But there will also be plenty of innovation.

Starbucks has lost its way. Forcing people to buy something isn’t the answer. But I digress… 👇

This. This is the right answer.

Preach.

The opposite of depression is not joy, but expression. Creating is the antidote. My hope is that AI, in time, lets us create more than we ever imagined, verses the dark depressive alternative.

Proving once again that, if at all possible, we must regain or retain that child-like wonder.

“It took me 4 years to paint like Raphael, but a lifetime to paint like a child.” —Pablo Picasso

homer simpson from the simpsons is standing in front of a grassy field . ALT: homer simpson from the simpsons is standing in front of a grassy field .

If the statement is true, I believe this is most likely a Ctrl-Z.

For sure. I’m reminded of the MAYA principle. Most Advanced Yet Acceptable. Many of these are advanced, but are they acceptable? In time, I believe we’ll get there. Google Glass and others were doomed by their timing and the expectations of the market. Those expectations will shift eventually.

I love the notion that we start by not changing our customers, but letting our customers change us.

I wouldn’t say Siri is for conversation, rather more commands. While they only recently added the ability to chain commands, it is not even close to what you can do with ChatGPT. It’s kind of embarrassing.

It’s clear Orion is the most logical (and useful) form factor for XR in the future.

The Vision Pro is a technical marvel, but it’s also impractical for daily use, which is surprising coming from Apple.

100% this. Forget evergreen jobs to be done or customer outcomes. Business acumen is one thing, business at the cost of customer “value” is another.

AI models collapse when trained on recursively generated data - Nature  Analysis shows that indiscriminately training generative artificial intelligence on real and generated content, usually done by scraping data from the Internet, can lead to a collapse in th...

We know what happens when the models start consuming their own synthetic content. Perhaps this is a good thing in the long run. Ha.

With the right checks and balances, it’s efficient and leads to more motivated teams and better customer and business outcomes.

Eilon Reshef Gong CPO Eilon Reshef on their unique team pod structure, how pods work with design partners, his spiral method for mastering complex topics, and more

Why again is this a unique structure for software development teams?

Perhaps we’ve shifted away from autonomous teams empowered to move outcomes together because of “efficiency.” But this, to me, is the optimal structure for cross functional digital product teams.

So, why can’t we add controls to control center on WatchOS?

I just wanted my Nationals Soto jersey to be wearable again. Doubt Lerner will go that high.

Minority Report was ahead of its time.