laura suttie

I wanted to wrap up this week by finally sharing my personal website! This site is part portfolio, part experiment, and part building with AI. Built with Codex and GitHub Pages, which meant I was the PM, designer (generous), copywriter, and confused engineer.

My baby loves to crawl and tackle me, we play peekaboo, he cracks up when I wear glasses, he loves watching me go up and down the stairs, he scrunches his nose when I whisper in his ear, he smiles so big with his toothy, adorable grin. I know there is no way to truly record these moments. The act of capturing them would, by definition, alter them. But I wish I could. So I go to sleep every night holding onto these memories in my head, with the hope that in 10, 20, 30 years, I’ll be able to replay them again. ❤️

i have lost the first two fantasy football games of the season (argh!) i asked @Muse to help me optimize my roster. i am determined to win this week's game. so far, it has recommended i pick up the available player with the highest projected points and didnt make any changes to my active roster. i had to prompt muse to review my defense, and evaluate my team more holistically... and it just isn't getting the strategy. focusing solely on projected points. (ex. do i keep nabers given dart is out?, do i pick up the panthers defense even though they are boom and bust?) but given i havent won a @ESPNFantasy game yet... maybe i should take muses advice...?

Made tiramisu for my husbands birthday tomorrow! First time making it 🤞

Sometimes the highest leverage product decision isn't what you build. It's what you decide not to build. A few weeks into one of my previous PM roles, I recommended shutting down a huge project. Not exactly what I expected to be doing. Aren't PMs supposed to build stuff? But the project had already consumed a huge engineering investment with no clear user outcome, no success metric, and a product integration that was barely functioning. So I dug in: * Researched alternatives * Compared product gaps * Estimated the engineering investment for each option * Looked at the roadmap and opportunity cost * Documented the tradeoffs My recommendation: stop the project. It wasn't popular at the time. Looking back, everyone agrees it was the right call. One of the hardest parts of product is recognizing when you're falling into the sunk cost trap. And in the age of AI, I think that judgment may matter more than ever!

Resolve ambiguity

Something that drives me crazy about ChatGPT: It’ll suggest a block of text, then say, “But if you really want it to be good, I’d do X, Y, Z.” Well wtf. If you thought that would be better, why didn’t you just do that in the first place?

The map on my personal website was the feature that took me the longest to get "right." I wanted to share places that have shaped me or left a lasting impact, and I "thought" it would be pretty "straightforward". I started with a static world map and manually placed a pin for each meaningful location, with links to relevant places in a popup. But lol... this broke pretty quickly. Since the map wasn't using proper coordinates, it wasn't responsive. A pin that looked correct on desktop would be way off on mobile. I then explored two options: • Google Maps • Leaflet + OpenStreetMap I decided to go with Leaflet. It solved the geography problem while giving me more control over the design. And introduced a whole new set of design questions: • Making the map responsive. Do I show the whole world? Do I let people zoom? • Popup sizing. A popup that's too large gets cut off or becomes hard to close on mobile. • Balancing pin size with tap targets in smaller, more crowded places like New Zealand. I couldn't stop making tweaks. It was addicting, and honestly, I still don't have it perfect. Definitely one of those features that seemed simple until I actually built it. :)

ok for people have their babies on a "nap schedule". do you wake them up from naps?? and do you sleep through the night??? asking for a friend...

website: laurasuttie.com Let me know what you think!

I think one of the most uncomfortable parts of this job market is pushing myself to follow up or reach out again (and again). Most of the time it doesn't change the outcome. But here's hoping for the one time it does :)

My sweet baby is sick ☹️ and as I’m sitting here with him asleep on me in the carrier, feeling for a moment a little trapped, I’m also reminded that I am completely irreplaceable in this moment.

Me, myself, and Claude

Codex keeps asking me to approve things as if I know what I’m approving… I don’t. So I responsibly click “Allow once.” How much damage can one approval really do? 🙃

my husband just said to me for the third time this evening: "just don't over think it". oh yeah what a great idea

5 lessons from trying to vibe code: (1) There’s still so much value in writing out exactly what you’re trying to build and why before you start building. You cannot outsource thinking and creativity. (2) I’ve spent a surprising amount of time trying to optimize API calls so I don’t accidentally blow through a hobby plan and spend $$$. (3) Design is still really hard. (4) I finally understand why engineers spend so much time perfecting tiny UX interactions. It is SO frustrating when something doesn’t work exactly the way you want it to. (5) Maybe I’m doing it wrong, but it’s still not that easy to get started building. To be able to ship something, I still need to know a lot about how github/code/hosting etc works. But when you do… wow. What a feeling.

Hmm……