Kenneth Aasan
@kennethaasan
Followers
401
Following
46K
Media
315
Statuses
5K
Senior Staff Engineer, @sportradar.
Trondheim, Norway
Joined April 2008
Better frontend output starts with tighter constraints, visual references, and real content. Here’s how to build intentional frontends with GPT-5.4 https://t.co/6Jgwcmqd9f
developers.openai.com
Practical techniques for steering GPT-5.4 toward polished, production-ready frontend designs.
127
209
3K
sent this to the team today everything great comes from being able to delay gratification for as long as possible and it feels like we're collectively losing our ability to do that
254
707
7K
me giving all my company private data to claude just to center a div https://t.co/FK38AIqhgj
44
303
7K
Over 80 years since Norway last had a women’s Allround champion… and today history was made. Ragne Wiklund 🇳🇴 brings the title back to Norway! 👏 👊 #SpeedSkating #WorldSpeedSkating
4
78
505
Introducing the Google Workspace CLI: https://t.co/8yWtbxiVPp - built for humans and agents. Google Drive, Gmail, Calendar, and every Workspace API. 40+ agent skills included.
github.com
Google Workspace CLI — one command-line tool for Drive, Gmail, Calendar, Sheets, Docs, Chat, Admin, and more. Dynamically built from Google Discovery Service. Includes AI agent skills. - googlework...
655
2K
15K
Ahhhh, Codex 5.3 (xhigh) with a vague prompt just solved a bug that I and others have been struggling to fix for over 6 months. Other reasoning levels with Codex failed, Opus 4.6 failed. Cost $4.14 and 45 minutes. Full trace plus includes original issue:
ampcode.com
120
231
4K
i love the drizzle team and i'm so happy they're joining planetscale many years ago i came across their team in my replies mentioning their tool they'd been working on it for a while but it didn't have much traction, didn't think much of it one day i did a twitch stream
30
31
2K
We have reached 300k users who regularly use their Codex sign-in through opencode. Way to go @thdxr and team. Congrats!
66
32
2K
On one end, the Anthropic team is a massive user of AI to write code (80%+ of all code deployed is written by Claude Code). They ship amazingly fast. On the other hand, seeing these beyond terrible reliability numbers suggests there might be a downside to all this speed:
307
269
5K
CTOs after their entire platform gets vibe coded over the weekend:
37
16
578
Plumbers and electricians seeing AI replace everyone who went to college
757
5K
40K
@elijahmuraoka_ I work in waves. Some are intense, some are less so. Been years where I tried to push the "normed" way onto me but that's just not me. I am far more successful with cycles of high/low output.
80
36
2K
If you're a Norwegian athlete of any kind, you'd want to look up if there are any events in Italy this year that you can sign up to or qualify for.
9
42
2K
It reminds me of this episode from Silicon Valley.
We’ve identified industrial-scale distillation attacks on our models by DeepSeek, Moonshot AI, and MiniMax. These labs created over 24,000 fraudulent accounts and generated over 16 million exchanges with Claude, extracting its capabilities to train and improve their own models.
16
196
2K
Silicon Valley was so ahead of its time.
We’ve identified industrial-scale distillation attacks on our models by DeepSeek, Moonshot AI, and MiniMax. These labs created over 24,000 fraudulent accounts and generated over 16 million exchanges with Claude, extracting its capabilities to train and improve their own models.
61
646
7K
Bizarre that I have a mental model of who Codex & Claude are and it’s basically these two
89
278
5K
This explains why the only major model that still suck at tool calling is Gemini
We’ve identified industrial-scale distillation attacks on our models by DeepSeek, Moonshot AI, and MiniMax. These labs created over 24,000 fraudulent accounts and generated over 16 million exchanges with Claude, extracting its capabilities to train and improve their own models.
63
46
3K
the irony
We’ve identified industrial-scale distillation attacks on our models by DeepSeek, Moonshot AI, and MiniMax. These labs created over 24,000 fraudulent accounts and generated over 16 million exchanges with Claude, extracting its capabilities to train and improve their own models.
15
26
517