@JamesMalsawm

Exploring AI by building real things. AI Tools, Vibe Coding, Indie Hacking. Sharing what I learn along the way.

Joined October 2022
  • Tweets201K
  • Following773
  • Followers5.6K
862Media
Argon + Opus 5.5 is a killer combination. one crunches data, the other builds visuals.
1
2
45
is this $4.99 Google Gemini ai pro deal actually worth it? does it even get Gemini 4 Argon?
15
1
22
472
I don't understand. DeepSeek is cleary win in price. why is everyone still paying $200 for Claude?
16
21
1,414
bro to bro.. we are 17 days Tibo reset in. did you actually do the daybreak blue setup yet?
Tibo
@thsottiaux3h
Day 17/ Bet you thought we forgot, but no, it’s reset day today. Also good time to remind you to get Daybreak Blue access, it only takes 15 mins and you don’t want to have to do it in the middle of an incident.
9
12
524
Which one runs your vibe coding? A. Opus 5.5 B. Haiku 5.5 C. Sol 6.1
25
1
29
648
Devin Fusion beats Claude Code Opus 5.5. $9.01 vs $13.04 per task. Is Opus 5.5 really getting nerfed?
9
15
400
5x the price.. claude $20 ≈ 2.5B opus tokens, codex $100 ≈ 3.2B sol. which are you paying?
19
2
25
565
wait.. did Claude just kill ChatGPT Dots before they even took off? 👀
Tibo
@thsottiaux12h
Dot, Dotter, Dottest
9
1
18
532
I didn't expect this.. sycophancy rank: sonnet 5.5 #30, argon #81. blunt or agreeable?
8
1
14
602
bad news guys: 80% of sol-only codex quota billed as astra. phantom or clean?
10
1
15
443
devs are asking if fable 5.5 is even happening. still waiting, or settled on opus 5.5?
13
17
649
claude security re-scans your repo as code lands. enterprise only. claude reviewing your code, or you?
2
5
360
I can't unsee this: anthropic says today's models are helping train future ones
2
7
313
wait.. openai just put codex and gpt-6.1 sol on google cloud. enemies hosting enemies
8
13
299
what the hell.. claude duet runs opus 5.5 while you sleep. my limits won't survive
5
5
257
gpt-6.1 sol subagent reasoning traces look like this. good luck auditing that
5
9
387
guys.. nvidia is selling whole ai factories now. is compute still the bottleneck?
3
3
285
I think claude code making a gta mission about vibe coding is peak 2026
15
12
422
sycophancy chart: grok 4.7 flatters least, gpt-4o flatters most. explains a lot
9
10
471
php sites.. openai's models found critical flaws in php. patch tonight, not monday
6
14
452
Load more