Steeve Lennmark

@xbxnxdxcxtx @xai I built dnsvard (OSS, dnsvard.com ) while building my SaaS. Not AI (like my SaaS) but pretty cool tech. Portless done slightly "righter"

@thdxr Yay we welcome you back to the bright side!

i thought i was out i stayed out of the game for 7 years but here we go again

@prisma Please write SQL instead. This is not the way forward.

Over time we kept running into the same problem with Prisma ORM. The architecture that got us here started slowing us down. Fixes took longer. Adding new capabilities got harder. So we asked: what would Prisma look like if we rebuilt it today?

@vineerpasam Good luck!

You wake up and see a vibe coder cloned your entire app. Same idea. Same features. What’s your next move?

@NehraWorkss It's ultimately bad for the users who will have their credentials leaked, their privacy violated, their money stolen and so on

Vibe coding really helped a solo founder to build product and ship it. what are your thoughts on this is it good or bad in long run?

v0.1.4 of dnsvard is out 🌐 dnsvard handles local dev networking end-to-end: - DNS + HTTP (80/443) + TCP routing - runtime + Docker visibility: dnsvard ps - fast cleanup by pattern: dnsvard rm -f <pattern> Built for multi-workspace workflows

@wesbos I prefer running dev myself for each branch/worktree and let the LLM know that the frontend reloads automatically, that way it doesn't have to spin it up itself

Port 3000 is in use, trying another one... Port 3001 is in use, trying another one... Port 3002 is in use, trying another one... These LLMs love to run multiple instances of the same app. made a Vite plugin `killer-instincts` that will either return the process ID of what is running on your strictPort, or kill it automatically

@swyx @ankitxg Let the slopacolypse begin. 🍿🍿

this is the Final Boss of Agentic Engineering: killing the Code Review at this point multiple people are already weighing how to remove the human code review bottleneck from agents becoming fully productive. @ankitxg was brave enough to map out how he sees SDLC being turned on its head. i'm not personally there yet, but I tend to be 3-6 months behind these people and yeah its definitely coming.

Thanks a lot @useblacksmith. Easy quality-of-life win, my builds went from 3.5m+ to <1m

I just shipped dnsvard v0.1.3 🎉 Zero(ish)-config local DNS + HTTP/TCP routing for multi-workspace dev. Run parallel branches/projects with stable hostnames, not memorised ports. Works with containerised and plain local services. Feedback welcome: dnsvard.com

Imagine running for your life, and this is your top speed 😂

@JustJake Sorry but garbage take.

The implementation no longer matters People really, really need to understand that

@MrPitbull07 $40 is too much on a $500 bill. We can't base tips on percentage, it's good for mental math on smaller amounts but doesn't work on bigger bills. The tip should be based on performance but obviously that has some kind of ceiling.

We went to a dinner as a group and had a $500 bill. We tipped $40. We were happy we can be able to give our server something, but her reaction was the opposite. She told us she assumed we're going to give her at least $120. When we asked for the manager, she said she was just joking, but she wasn't smiling at all. Idk, but is $40 tip enough for $500 bill? I just feel like expecting $120 is not realistic. ~Lea Robertson

@ryusudol Does anyone use anything else at this point?

Does anyone use Bun? What’s its strength?

@willvincent @brankopetric00 It's not an opinion, just fact. Compose is not a production component. How do you roll out new releases without downtime? It's fine for your private blog with no visitors but not for any real production environment

@drblard @brankopetric00 Your opinion, while noted, is not fact. 👍

Management wants everything on Kubernetes. You currently run: - 3 EC2 instances running Docker Compose (production) - 1 RDS PostgreSQL database - 2 background workers on dedicated instances - A cron server running 12 scheduled jobs - Total infrastructure cost: $1,800/month - Team size: 4 developers, 0 dedicated DevOps - Deployments: SSH + docker-compose pull (takes 2 minutes) Your application: - 6 services total - 400 requests/minute at peak - 99.5% uptime target - Simple scaling needs (horizontal on the API only) The proposed EKS cluster would cost an estimated $3,400/month. Do you migrate to Kubernetes, push back on management, or propose an alternative? What's your argument?

@willvincent @brankopetric00 It's not for production. Simple as that.

@drblard @brankopetric00 I can parrot your reasoning back to you and be correct, if setup properly it (docker compose) also *just works*

@kr0der Raw git, always have, always will

what git client do you guys use alongside Cursor? lazygit? github desktop? just raw git commands in the terminal?

@garrytan If I'm to believe people on X that's half a days worth of commits.

We are so back

@alfredapp Yo A! I tried raycast the other day (sorry). I couldn't stand it though and switched back to my trusty Alfred. After the switch-back I see random delays of up to about a second for Alfred to appear, you're punishing me because I was unfaithful? :-)

@LewisJEllis More like neovim vs emacs

a year ago: cursor vs windsurf now: claude code vs codex a year from now: linear vs conductor