Filled our bellies in Singapore and have now landed in Malaysia, living and working with this view for the next few weeks.
English
Chris Halaska
9.6K posts
Chris Halaska
@chalaska
I left Google to build the studio I always wanted to work with. Product direction, strategy, and design for founders.
Katılım Nisan 2009
3.1K Takip Edilen13.8K Takipçiler
My son is only 21 months old and is about to be in his 8th primary residence.
Good experience I guess? 😅
English
Time spent recording looms - decreasing
Time spent prompting replit - increasing
English
AI is excellent at gathering research. Mapping competitors. Clustering themes across interviews.
It's terrible at knowing which of those findings actually matter for this specific product, at this specific stage.
English
I used to spend two weeks on discovery before a project could move forward.
Now it takes days.
But the time savings aren't the interesting part. What AI still can't do is.
English
Exploring Singapore today with my fam and enjoying all the simple things like this chicken rice
English
At Google I learned that "looks clean" isn't a strategy at billion-user scale.
At CBA I learned that trust signals in banking are completely different to consumer tech.
Running my own studio I learned that the clients who pay a premium are paying for taste, even if they never use that word.
English
Had enough of the daily emergency alerts in Dubai, heading to Singapore for a week or so to make sure my family is safe and then will see!
English
Interesting challenge designing an AI product:
The AI design tools hit their limit when you need complex, smooth interactions.
They can generate components and basic flows, but nuanced interaction models? Still need designers who understand the craft.
AI speeds us up. Doesn't replace taste.
English
Three design problems every AI product faces:
1. Trust: users don't believe the output
2. Focus: too many capabilities, no clear starting point
3. Feedback: users don't see the AI improving
Solve these three and you've solved 80% of the experience.
English
Pre-mortem exercise for AI products:
"It's 6 months from now. Your product launched and users aren't coming back. Why?"
Do this in week 2 of your sprint. You'll redesign half the product based on the answers.
English
Chris Halaska retweetledi
British startups are increasingly using artificial intelligence and freelancers instead of taking on permanent staff, per Bloomberg.
English
The most important screen in your AI product isn't the results page.
It's the error state.
That's where trust is built or broken.
English
Taste isn't "does it look good."
It's "does this serve the user, the product, and the business at the same time?"
That's the skill AI can't replicate. And most designers don't invest in building it.
English
Designing for AI has made me rethink everything I thought I knew about empty states.
An empty state in a traditional app means "you haven't done anything yet."
An empty state in an AI product means "I don't know anything about you yet."
English
AI makes production cheaper and faster.
That means the market will be flooded with "good enough" design.
The designers who stand out won't be the ones who produce the most. They'll be the ones who consistently make the right calls.
English