@grok is not even close to opus on coding. When will someone catch up with anthropic?
English
Timothy Wright
600 posts
Timothy Wright
@timothywright
regular guy trying to look busy whilst accomplishing very little
Boone, NC เข้าร่วม Eylül 2022
613 กำลังติดตาม143 ผู้ติดตาม
not being able to use anthropic models in opencode has really slowed me down. I have realized that I am doing more data cleanup than I thought, and kimi/other stuff is not even close to anthropic on this.
English
@DanielLockyer The phrase "the merchants of complexity" goes through my head about 100 times per week as I navigate what aws and a team of overseas contractors have convinced my boss to be the way.
English
@johnlpollard That sounds like a good open source contribution for agents.md
English
@timothywright "Do not write code like John did previously. Past John was an idiot who thought he was smarter than he really is"
English
This makes me think Im doing it all wrong. Instead I wonder if I should keep a local copy of fizzy and tell AI "refactor this controller so that it looks like one from fizzy". Now AI could have a much better context than even I might be able to communicate.
claire vo 🖤@clairevo
"I avoid prompting at all costs." That's @jameygannon, a brand designer and AI creative director who has spent thousands of hours in @midjourney building a system that lets her create consistent, beautiful brand imagery with AI -- without complex prompting and srefs. In this ep you'll see: - how @Pinterest mood boards communicate your aesthetic to Midjourney better than text - how to build a personalization code so Midjourney learns YOUR taste specifically - her @floraai flow for fixing and delivering final assets to clients This is by far the prettiest episode of How I AI, and has already changed how we make thumbnails for the podcast. Thanks to our awesome sponsors 🫏 @TrustVanta - Automate compliance and simplify security 💜 @Lovable - Build apps by simply chatting with AI Watch now: youtu.be/2RD3FP5iWJY
English
@nateberkopec For moderately dynamic pages, rails cache or cdn?
English
I'm interested in helping people with Rails apps w/significant #s of public-facing pages to support more HTTP caching.
The best path forward here is to move as much as possible to the CDN and let them deal with it.
Marc Köhlbrugge@marckohlbrugge
AI crawlers are out of control 326 requests per minute??
English
i stopped going to menswear forums the day i realized i couldn't think of anything i've learned there in the last five years. i've since started wondering what i've learned from reading social media and i honestly can't think of a single meaningful thing.
Bryan Johnson@bryan_johnson
Finished a seven day social media fast. It feels like the most effective longevity therapy I've done. Everything got better: mood, sleep, energy, presence, judgment, relationships, and optimism. Evidence shows a seven day fast produces a reduction of anxiety (16%), depression (25%) and insomnia (15%). The effects felt bigger. Conversely, dipping back in, I can viscerally feel that my body metabolizes social media similarly to a fast food meal, corrosive relationship, hangover, and sleep deprivation. My body hates it. After the previous fasts (40/hr and 70hr), I wrote that social media is pollution. Not a vice or guilty pleasure. It’s closer to water toxins, air pollution and microplastics. This time, the major insight was that social media is a form of intoxication. Alcohol is honest intoxication. It clearly tells you what it's taking from you. Social media on the other hand does not disclose itself as an intoxicant. It produces the sensation of being informed, engaged, and connected while quietly evacuating your capacity for depth and independent thought. You don’t feel drunk, you feel current. But evidence shows that it causes your brain to shrink. The impairment is real by you can't feel it. Making it the more dangerous type. If you haven't tried it, I strongly encourage you to try a social media fast. Even if for one day.
English
@cgtwts I had a cloud storage bill go from $80 to $10k with no explanation. They declined to adjust and also declined to tell me why the traffic was deemed as legitimate.
English
@zeeg You are a merchant of complexity and can't see any other way.
English
@timothywright More people are misinformed by him than me?
Nuance is dead
English
Most of us have enough customers that our logs don’t live on a single node 🤷
Or we have you know SLAs or other real world concerns
@levelsio@levelsio
Let's pay $99/mo to read my logs How about TAIL??? tail -n 50 /var/log/{php*.log,nginx/error.log,syslog} 2>/dev/null | claude "analyze these logs for errors or issues"
English
@zeeg More people listen to him than they do you. More builders are inspired by him than they are by you.
English
There’s a growth mindset people like this guy don’t have
“I don’t need solutions for my problems because my problems will never be big enough to require them”
That’s fine, but my man you do not represent real technology companies or businesses
English
@levelsio I hooked my agents up to fizzy and interact with them there. Adding notes to cards and assigning stuff. Commenting back. Openclaw orchestrates it. Wouldn't be a stretch to pipe error logs to it
English
This is fun but what if we connect my PHP, JS and server error logs AND the bug and feature request board to an AI that then writes the code and does pull requests that I then approve or reject?
How I do this? Can I do it with Claude Code?
@levelsio@levelsio
This week I decided to just permanently switch to running Claude Code on the server mostly on bypass permissions mode: c() { IS_SANDBOX=1 claude --dangerously-skip-permissions "$@"; } And for the first time in my life I think I've actually managed to outrun my todo list What happened is I simply blasted through my to do list of features I had to build and bugs I had to fix I've never shipped so fast and Claude Code almost made no mistakes, and when it did it they were tiny that weren't fatal (important because I'm mostly working on the server in production now) Before I was always known to ship fast (also because I always work alone) but while I shipped new things would always build up on my features/bug board (my users can submit them there) But this is the first week where I've been fast enough to outrun them The board is actually empty! As other people have written on here the real bottleneck is becoming myself and my creativity, not how fast I can ship. Because I think I ship faster now than I can come up with new ideas, or maybe my brain will adjust to this new speed (probably) Also I feel another limit is becoming my own mental context window, as in how many things, features, bugs, projects, I can keep in my mind in parallel while building on all of them. It's a lot and I haven't reached that limit yet but I feel I might be close I also noticed that you start going really fast the more you let it just go loose, before I was slow because I didn't trust it and I was scared it would destroy my code, now I just let it go. As @karpathy wrote, things feel like they've changed a lot around December last year when models became good enough to really code with and I feel the same When I see other friends code with Claude Code I often notice they're slow because they still check everything, which is good of course, but I feel the better way would be to create some tests and just let it run freely and see if it can pass those For me the tests are mostly just me checking out if the new feature on the site works or not, and in 99% cases it just does, and then I ask it to improve it further Because I run Claude Code on the server in production, I don't have to wait for deployment anymore (although that took only 3 seconds anyway before, that still adds up), now it's wait for it to be done coding, I refresh the site and I test it, that feedback loop is how I work and it's made me WAY faster Anyway here's what I did this week and the majority of these things were requested by people on the bug board, I'd say this is about 10x my normal output: 📸 Photo AI - Built new image viewer and mobile image viewer - Added batch remix, multi-photo import, filtering by model in gallery - Security overhaul: phased out insecure ?hash= login, migrated to session tokens - Fixed Google login loop, multi-model selection, talking scripts - Added custom audio upload for talking videos - Created dynamic model selector from server endpoint 🏡 Interior AI - Revived [ Add furniture ] feature (started 6 months ago, image models now good enough) - Added custom style upload for redesigns - Built own Gaussian Splat viewer for 3D - Made /remove_bg endpoint for furniture backgrounds - Migrated 3D walkthrough to new World Labs API - Added .skp file support, paint color masking, empty room button 🎒 Nomads - Launched weekly AI-generated newsletter from chat - Built profile edit modal, moved profile editing from /settings to profile page - Added TikTok/YouTube links, status bar, server-side API tracking - Added hundreds of new profile tags and traits - Fixed timezone filters, broken links, user avatars 🗺️ Hoodmaps - Revived write mode (before was only read for last few years because db was rekt) - Built heatmap mode using sentiment-scored tags (50K+ tags) - Fixed root cause: tags not entering DB due to wrong PRAGMA (should be WAL) - Added good/bad area detection with admin grid controls - Set up Claude Code Telegram bot for live changes - Enabled CF cache, fixed health check, fixed Brussels 📕 MAKE book - Built auto ePub/PDF generator cron worker - Added dynamic generation with personal customer watermarks - Added image compression for file size 💾 Pieter .com - Added Wikipedia text-only reader for Kindle - Exploring Windows 3.11 emulator using v86 (to replace Em-DOSBox) - Added product recommendations on homepage - Installed Wall Street Raider (1986) 👩💻 Remote OK - Installed Chatbase AI customer support bot - Added "report not remote" link on job posts 🏨 Hotelist (3 todos) - Fixed hotel URLs and city range bugs - Added iron amenity
English
my company uses jira, but I hate it! I use a workflow where I screenshot issues in jira, feed them to a google chat with openclaw, and it creates the issues in Fizzy for me, and I work them from there. When done with an issue, I grab the last comment that openclaw left, which is a summary, and put that back into Jira. Last step when things are done is having the integration I have with writebook from my rails app update our documentation. @37signals helping me not use Jira, even when I have to use Jira.
English
daily scrum meetings are a wild way for people who need validation to get it.
English
@sysadafterdark Windows xp sp3 and server 2008 r2 were strong.
English
The best version of Windows. We still talk about you.
Computer ♥ Records@ComputerLove_
On this day in 2000, Windows 2000 was released.
English
I just spent ages on a fairly simple issue with what i thought was opus 4.5. After about a dozen attempts with stupid errors, I noticed that opencode model was set to gemini.
English
As part of today’s repeal of the 2009 Obama EPA Endangerment Finding, the Trump EPA is also ENDING all off-cycle credits, including…🥁🥁🥁...the almost universally HATED start-stop feature in vehicles.
Fixed it!! Promise Made. Promise Kept.
Lee Zeldin@epaleezeldin
Start/stop technology: where your car dies at every red light so companies get a climate participation trophy. EPA approved it, and everyone hates it, so we’re fixing it.
English
@dieworkwear who knew that a thread on cotton blends was how my day needed to start? But here we are!
English
Interesting to see people make LLMs of you because you know what you would say and you can see the LLM is wrong. This is harder to judge when you make LLMs of other people bc you have nothing to judge the accuracy against.
I will tell you my view of 90/10 cotton-synthetics 🧵
Rui Sousa@HeyItzaMi
Derek Guy @openclaw agent is already paying off with the knowledge output 😄
English