Ron
@ronzilla
Building: Try That LLM. Using LLMs via API? Wondering if you should change models? Try That LLM can help.
You might like
I'm beyond excited to announce Try That LLM: a service for people using LLMs via API. Bulk-test your project's prompts against dozens of LLMs, automatically test them against every new LLM as it ships, and set up LLM judges to score the outputs. Try That LLM is designed for…
I guess I've never done the math on what it would take to create your own video streaming site, but I wouldn't have guessed "one day to code, 80/month to host"
🌶️ seoroast.com too spicy for YouTube Channel flagged. 1k subs, 107 videos, gone Vibe-coded a video platform in 1 day: - YT Takeout → @FFmpeg → HLS stream + Nginx - Transcripts with Whisper - Clickable chapters, custom video player - No ads, no algo, €80/month!
Can I just say that in this month of "Best TV shows of 2025" and Best Movies of 2025", justwatch.com is a lifesaver.
Is the right metric "code was AI-generated or not", or is it "here's my confidence level in this code"? A "I'm less confident about the change to this file" would be handy for human changes too.
Unpopular opinion: Current code review tools just don’t make much sense for AI-generated code When reviewing code I really want to know: - The prompt made by the dev - What corrections the other dev made to the code - Clear marking of code AI-generated not changed by a human
This it totally true, and applies to so many areas: careers, scientific breakthroughs, dating, etc. We rarely recognize our turning points in real time. They do not introduce themselves. Sometimes they look like chance. Sometimes they look like discomfort.
Grateful to be featured in the @StanfordDaily - the first part of the Beyond the Farm series. Thank you, @diyasabh, for telling this story with such generosity. Reading it felt a little like opening a time capsule I didn’t realize I had been collecting: old stories from post…
I watched the Seattle Torrent women's ice hockey game on Friday, sitting among the record-setting crowd of 16,000 in the same arena that the Seattle NHL team plays in. Rest assured the chants of "Ref you suck" are just as loud at women's games. It was super fun, a great…
This is a screaming deal--honestly the Find Your Following sample videos sealed it for me.
Build a product. Build a business. Build an audience. I highly recommend buying the bundle before it jumps from $25 back to $100+
[Redacted Protocol] is an outstanding company name
The classic World Flags Graded still has the best flag grades: worldflagsgraded.com/ratings.html
No politics attached. What are the best looking country flags from a purely visual standpoint? Here’s my two favorites.
A parody account that's tied to a company's mission...with great comedy writing to boot. Nice work @compliantvc.
This startup got 10M impressions in 15 days using a… parody account! Henrick doesn’t exist. He’s a fictional character of Comp AI. But the brand awareness he drives is very real. He posts hilarious content about rules, regulations and the EU’s obsession with them. Not…
Those are the worst! "It's just a string change in a dropdown, why would it break anything?" Is there a better alternative than E2E tests? With the inevitable flakiness wrangling...
Oh boy probably just had one of the bigger fuckups for HelpKit in years 🤦♂️ I nuked the entire onboarding flow of HelpKit for 2 full days. Nobody could signup for the trial anymore. Here's what happened and a lesson: Someone told me I should add ChatGPT as one of the discovery…
"Come up with a few solutions for this" and "Don't write any code yet" are my personal favorites.
asked 7 people at @cursor_ai how they use cursor and if they have any unconventional workflows. everyone works a little differently, from tab to fully autonomous background agents. some replies:
This is a super cool idea: transform your git history into marketing materials and release notes, via a cursor command. The prompt could be tweaked to answer big-company questions too, like: - a weekly summary for a manager - internal notes for other teams
Pollo is pretty nice--their video model selection is excellent!
WAN 2.5 is now LIVE on Pollo AI! 🔥 Now with synchronized sound, cinematic camera control, and smarter AI understanding for stunning 10s videos. To celebrate it, all PAID USERS get 50% OFF Credits until September 30th! 🎁 24-HOUR RETWEET GIVEAWAY 🎁 - Everyone who Retweets…
I'd also like to know what the LLM thinks are the best of my phrases, across my various coding conversations. Like I know I use a lot of "Great!" when things worked, "Hmmm" when things didn't, etc.
we should start a collection of the best LLM phrases. i’ll go first: Ah, I see the issue now
United States Trends
- 1. Happy Black History Month N/A
- 2. Mike LaFleur N/A
- 3. Raheem Morris N/A
- 4. Nate Oats N/A
- 5. Solanke N/A
- 6. Jordan Hicks N/A
- 7. Todd Golden N/A
- 8. Sandlin N/A
- 9. Gators N/A
- 10. #GRAMMYs N/A
- 11. Florida N/A
- 12. Devers N/A
- 13. Azzi N/A
- 14. UConn N/A
- 15. White Sox N/A
- 16. Charles Bediako N/A
- 17. Rodri N/A
- 18. Kansas State N/A
- 19. Rueben Chinyelu N/A
- 20. Gage Ziehl N/A
Something went wrong.
Something went wrong.