ai is one helluva tech bubble. things are going to get pretty crazy when it pops

Involve the team in creating the strategy. And in setting the agenda for the next quarter. So then you can let go of how they work their magic, but they aren’t being asked to do something impossible.

with all these ai agents taking our jobs, at least pepe’s still in work

“The power of AI lies in helping us write less, not more. Instead of asking AI to expand your thoughts into lengthy pieces, ask it to compress them.” codebytom.blog/2025/09/24/w...

i’m getting to the point where I’ve got chatgpt conversations that are so long my browser is crashing trying to load them

that feeling when a random ai notetaker joins a call

anyone can vibe code an app, but you still need strong research, product, design, and engineering skills to vibe code a good one.

People ignore formally-imposed metrics because they don’t believe it’s measuring something that’s actually important, or that they have control over. Another common reason is: They don’t want to face the truth of what the metric means. Both insidious; different solutions.

i’ve wasted so much time building and burning down productivity systems. what I’ve learned is exactly what @joanwestenberg.com says. personal productivity is all about context—what works for one silicon valley tech bro almost certainly won’t work for you. www.joanwestenberg.com/p/why-billio...

before you build anything, you need to make sure it’s the right thing—life’s too short for lines of code no one gives a shit about.

ai product management 101

I’ve been quiet on here, haven’t I? It’s been the holidays, though. And I bought a puppy.

had the iOS 26 beta installed for a few days. really enjoying the “quality of life” updates so far. not a major leap forward ux-wise, but definitely pulling me deeper into the apple ecosystem again

Whoever says, ‘as an engineer I can talk to customers and define product strategy myself, so we don’t need product managers’ just became the product manager

If you can’t write a user story and it’s acceptance criteria on a single large post it note, then it’s too big.

Social media is why the internet sucks now, but not for the reasons you're thinking. It sucks now because they convinced you to give up having your own website in favor of a centralized platform they control. This allowed them to normalize shit like identity verification.

had an excellent time talking at producttank newcastle again last night.

this is one of the main things that makes bluesky awesome

watching terminator 2 again. feels a lot more real now than it did thirty-odd years ago.

literally just done a talk on this exact topic

iOS could’ve been one of the all-time great gaming platforms, but Apple let the experience die in favour of advertising. When 90% of games mean wading through advert after advert before you even get to play, what’s the point in persevering to try and find something decent?

For most tech jobs, thinking is the bottleneck—not cranking widgets. That’s not a bad thing, though. If we’re not making room to think, we risk creating a lot of stuff that has little or no value. We should be using AI to give us that room back, not do the thinking for us.

The speed of Lovable’s growth has been incredible. As a PM whose coding skills amount to HTML, CSS and some VBA from back in the day, I’ve been blown away with how easily I can turn ideas into something tangible. Showing is better than telling, and Lovable has skyrocketed my ability to do that.

The Apple Notes meme is a pretty good representation of my note-taking journey. I don’t need a Zettelkasten or PARA setup, I just need to “do the thing” instead of procrastinating. With tools like Evernote, Roam and Obsidian I spent more time messing around with my notes than using them.

Just like GarageBand turned everyone into music producers, vibe coding tools are turning everyone into developers. That doesn’t mean what they’re developing is any good, though. When the world is flooded with half-baked ideas, it’s creativity, taste and experience that will make you stand out.

Product managers need to be able to sell. It’s all very well focusing on customer impact and achieving outcomes, but building software is a commercial endeavor. If you can’t sell what you’re shipping to someone, how can you expect your sales team to be able to?

Once we all start abdicating our thinking to machines we’re screwed. AI will fundamentally transform how we work, and how we measure our impact, but it can’t be at the expense of our own critical thinking. AI isn’t killing knowledge work—it’s making it more important.

Interesting to see how hard AI companies are starting to push into education to try and counter the “AI is evil and rotting our brains” narrative. Expect to see a lot more partnerships like the one OpenAI just announced with Canvas.

I followed @joanwestenberg.com’s lead and deleted my second brain this week—my reflect notes account, all my apple notes, all my reminders, my readwise account. starting again with a blank slate has been hugely liberating and has really got me thinking about what I’m capturing “knowledge” for.

Just because it's big doesn't make it strategic.

this is about video games specifically, but it's true of all product development in the age of ai. when you can build anything and everything, it's authenticity, integrity, and human-ness that will set you apart from the swathes of vibe-coded apps flooding whatever market you're in.

i was gutted when the browser company announced they were killing off arc, but now that i've played with dia for a few days i can absolutely see why. ai is going to completely reimagine how we interact with computers and the internet, and dia is already showing what's possible.

as a product manager, you need to try and embrace uncertainty—not eliminate it

simplicity is a competitive edge

not all meetings are bad, but bad meetings are the biggest killers of productivity and progress—avoid them at all costs

there’s a big difference between making a decision and acting on it

Build. Measure. Learn. Long ago, the three phases lived together in harmony. Then everything changed when the feature factory attacked. Only the Product Manager, master of all three phases, could stop it. But when the world needed him most, he was double-booked.

building digital products is nothing like manufacturing.

excellent post from @simplybastow.com on enshittification, why it happens, and what product people need to do to stop it.

vibe-coding a fantasy football scout app in the style of 1990s championship manager because, why not?

The biggest misconception people have is that leaders can solve their problems or unblock for them. I tell people to bring solutions/options not problems. Don’t complain, ask questions. Assume good intent. Create the condition for people to bring their best selves to the collab.

just realised I haven’t posted on here for a week. anyone miss me??

turns out having a humanities degree and a background in journalism is pretty useful when you’re talking to robots all day

"Don't tell me about your framework, show me what you've actually built." The last decade of product management was dominated by theory, but the pendulum is swinging back towards creativity. Best practices don't create value. What does are the products you can get into your customers' hands.

Current view on AI: profoundly interesting. At least a decade of opportunities and implications. AGI seems like bullshit. LLMs may plateau. I’m deeply uneasy about its impact on society and the IP and culture industries. Uncritical boosterism without parallel risk mitigation is a really bad idea.

should’ve just googled the train company’s help number. GenAI is super-powerful, but can be super-dangerous if you don’t know how it works or what its limitations are.

vibe-coded prototypes are the new prds

Both of these people will accurately state that AI has made them more productive and saved them time.

too many product teams mistake motion for progress. as a product leader, you need to keep asking how what you're doing is making a difference. in the post-zirp product era, a product team's success isn't measured on features shipped—it's measured on impact made, to customers and your business.

"what were the 2000s like?" they were just this. nintendo sold a tribal tattoo game console and it came with a winamp skin