When the Daily Was Actually a Stand-Up?
We used to stand up to keep meetings short. Then we went remote — and accidentally sat back down forever. Here's what happened to your daily, why it's costing more than you think, and why the plank might be the only thing that saves it.
When to Say No to a Stakeholder?
A book by Gabor Maté got me thinking about the PM interview question everyone asks. Here's how I say no to stakeholders — and how my approach changed over time.
When is the day that AI starts renting humans?
What happens when AI agents begin hiring humans instead of the other way around? This thought experiment explores a future where people become the low-energy, high-flexibility “hardware layer” for artificial intelligence — rented for physical edge cases machines cannot efficiently perform.
When AI Fails You: What to Do When Your Perfect Prompts Become Your Biggest Problem
I caught myself staring blankly at a simple email, reaching for ChatGPT. This is AI brain rot—and the better you get at prompting, the worse it gets. Here's the paradox nobody talks about and what to do when AI fails you.
Google Search, Prompt, or AI agents: when to use what?
Learn when to use Google Search, AI prompts, or AI agents as a product manager to make faster, smarter, and more defensible decisions.
When Will the Agile Manifesto Get an AI Upgrade?
Just a few years ago, our feeds were full of Scrum certifications. Today, AI is writing half our code. As we move from early delivery to 'instant' value, it’s time to ask: When will the Agile Manifesto get its AI upgrade? Dive into the new frameworks like AI-DLC and Vibe Coding that are redefining the triad of Business, Developers, and AI
When AI ate a cookie from Amsterdam? How to recognize?
How often is AI accurate? I dive deep into the numbers, from the best-in-class 99.3% accuracy rate to the shocking truth of confident lies. Discover what benchmarks mean and why you need to apply the verification tax to every AI output.
When will AI run out of internet? Is that the end of pre-training?
By 2027-2029 frontier AI labs will exhaust all high-quality public data. Here's what's next.
