Seema

@codyetc const antly

can't believe i used to type "const" all the time

I was always impressed by the MLB tv player and experience and it felt optimized for baseball watching. Little things like easily jumping to a specific innings, finding stats, avoiding spoilers if you started watching a game after it started. Not loving this ESPN switch thus far

AI coding tools may boost productivity but they also boost the damage bad engineers can do

Claude code just now: "Let me re-read the full file with fresh eyes." Uh...

@AGrillz @ycombinator this is legit impressive, nice infinite scroll :)

I added Founder profiles! Filter and search ALL @ycombinator founders. You can see: - Repeat YC founders - Career path (eng/product/sales/consulting) - Years work experience - Prev. notable employers - Education - College Dropouts Check it out --> yc-update-tracker.lovable.app/founders

I made a free web app that tracks @ycombinator companies. It updates when: - A new company is listed to YC’s website - A company does a launch post - A company changes their name or tagline/one-liner You can even download the full list of all the companies YC has ever funded.

- There is a certain ability to tinker that I find myself missing, might be the lack of visible code (you can see this but it's not the default) - I find myself wanting more transparency on the decisions made and the why behind them (why use certain technologies, patterns, etc.) - I can't help but feel like I could build a lot of what I want faster but I realize I'm not their target demographic (2/2)

Decided to try @Lovable with the free credits they are offering today. I feel like the thing it does well is the structured prompting. Being able to right good prompts is a skill and probably one of the challenges getting started with Open AI and Claude and whatever else. They make this more accessible. Some other thoughts: - I don't feel like the results are what I want– there is a certain craft missing in the designs and even iterating isn't really getting me there (1/2)

@AGrillz @codyetc seems directly correlated with your codebase contributions

@codyetc It me

found the bug

I asked a chatbot a question about my coffee machine after reading some comment threads (admittedly a bad decision) and it told me "I'd say close those comment threads and go enjoy the coffee." and to "go enjoy a decaf" 😵‍💫

I'm trying to buy sparkling water from @Costco and one of the suggestions is not like the others wtf

@united @romainhuet @united agree your app is generally great but it failed to notify me I was rebooked onto a new flight last week resulting in me spending the night in LAX. I was already at your counter waiting for an agent and checking the app. Can you help look into it?

@romainhuet Hi, Romain. Thanks for sharing this. We hope that you have a great flight! ✈️ 💙

One of the things I love about @united is how good their iOS app is. The latest version has real-time boarding with a passenger progress bar, smart lounge guidance based on your gate and capacity, and more detailed live bag tracking. Lots of thoughtful little details ✈️

@AGrillz @ycombinator @riveterhq F25

Demo Day for @ycombinator F25 is Wednesday! ✨ Reply "F25" for a link to a spreadsheet with the companies, their revenue, customers, and founder backgrounds. Made in minutes with @riveterhq (YC F24).

@frank_ I think you just described bias

i think about this often— if beauty is a subjective, evolved preference shaped by biology and culture. and if models learn and recognize patterns humans call beautiful via training data. then every time a model is recognizing something as beautiful, it’s effectively expressing the correlation between that something and the patterns it learned as being preferred by humans. models simulate the recognition of beauty through statistical correlations in human-labeled data. every time a model states that something is beautiful, it’s echoing the average of a thousand human voices who once said it was beautiful. that’s wild.

@MikeLyngaas @NotionHQ You can change it :)

Hi @NotionHQ , does anyone like dates in this format? 🤔

@AGrillz @ycombinator @riveterhq W25

It's time for @ycombinator W25 Demo Day already! 💬✨Reply with "W25" and I'll DM you the spreadsheet with details on all companies in the batch, made with @riveterhq (YC F24). Making this list took minutes in Riveter.

after 10 years of iterm2 and prezto i'm switching to ghostty and starship (with basically the prezto prompt ported over) 😵‍💫

@AGrillz i’m sure you can guess my answer 😄

🔩 Want to pull data from internal docs into Riveter? Now you can! 📝 With a simple prompt, extract details from PRDs (or any doc) into a table. Watch our quick video to see Riveter pull milestones, expected ship dates, project descriptions, and more into a spreadsheet in minutes. This demo uses Confluence - which integrations would you like next? Try it yourself with a free trial: cal.com/team/riveter/r…

DODGERS!!!! What a wild game and amazing victory! World champs!

my brother: what’s 300 divided by 25 me almost immediately: 12 rest of my family: (trying to figure out the answer) uhhh me: i told you guys it’s 12 my mom: i think it’s 13? me: … i can’t

@TessHuelskamp I sent someone logs for 9 am Pacific time when they wanted the ones for 6 am Eastern 😬