creator of @connect_Relay, https://xitter.cf/t.co/OQJ9nXoEy3 & Emc² Co-founder at https://xitter.cf/t.co/SIbh92djSp & https://xitter.cf/t.co/pt0TIQj2Nf

NOT in SF
Joined August 2020
Col update: 1,088 component entries across 69 libraries 🧩 • Added React Aria Components (54) @devongovett • Indexed all 213 free React Bits components @davidhaz • Fixed search mixing words from unrelated components collection-of-libs.vercel.ap… cc @itsluneri @fayequepeerzade
3
1
15
2,935
Now you can use Claude Mythos to keep your code base safe & secure. Devin Security 2.0 lets you use Claude's mythos & many top-level models for cybersecurity work. I tried it on my own app, Emc2, & it came back with some great additions for checks to do so that our users' data is safe & secure.
devin is giving access to claude mythos team just announced devin security 2.0 so users can use top level cyber models to keep themselves safe > it scans your code for bugs > checks which ones can actually be exploited in your setup > reproduces them, writes patches and runs tests > opens prs with fixes for your team to review team is offering free enterprise scans and credits for new customers to fix what they find with all hacks going on. it's really imp to keep everything safe and devin is here to us out
1
2
12
Since the launch of Opus 5.5, @AnthropicAI has been killing it & OpenAI has stayed silent. @thsottiaux is doing the 28-day challenge, but most of the days it's only Resets, which isn't very useful, as Astra is not usable at all & 6.1 Sol is mid & compared to Opus 5.5, both models suck. At this pace, @OpenAI needs to release 6.1 Astra, which can at least reach halfway to the Fable 5.5 Benchmarks & 6.2 Sol & Luna to compensate 😔😔
1
6
358
I know they changed something, but I just can't say it. Loving this new change tho
8
50
6,987
Maybe @rohit3a was right after all.... Light mode does look good. 😔😔 only on @gitcafe tho 😋😋
3
5
281
To dear @theo & @jullerino. Since the v2 is out now...are we ever going to get T3 code-managed skills? Or is that the plan for v3? 😭💀
1
2
36
Thank the lord that @DevinAI Security exists😭🔥 I didn't even knew this was a thing; I saw this yesterday at the @marclou & @DevinAI event! It really was worth it to go to the event!
1
7
266
My first ever meet & its @marclou + @DevinAI. Thanks @Avi31m & team!! Finally got to meet Marc!!!!! Tho I could not talk with him, listing to him was amazing, very kind & down to earth person. Time is always a constraint! Can't wait for the next one, Marc should come more often ngl!!
6
31
547
I was able to Connect with so many ppl it was great! For my first ever AI & meet event, this was amazing!
2
60
Overall, obviously there were some rendering issues & what not, but my laptop is not that great tbh. Will try this on another machine maybe. Also this is what Claude opus 5.5 made when I asked for a teaser.
I had @DevinAI Claude opus 5.5 medium + SWE 2 High make me a trailer for my upcoming game. "Nightmare Saga: Dream 1" by @ZnsStudios • Took 2.5 hours • ~10-12% of weekly (Had another thread running so not confirmed) • Ran into Audio & in-game asset issues
5
290
I had @DevinAI Claude opus 5.5 medium + SWE 2 High make me a trailer for my upcoming game. "Nightmare Saga: Dream 1" by @ZnsStudios • Took 2.5 hours • ~10-12% of weekly (Had another thread running so not confirmed) • Ran into Audio & in-game asset issues
3
1
2
25
1,073
The usage is not confirmed, I had another thread running on Devin for another project. There were some issues like the Doors being closed & like that where Claude could not do anything. It also was not able to add any Audio at all. Before this, Claude opus 5.5 made this vid.
1
2
23
This one was probably the best one in some cases.
1
3
9
The launch video for Opus 5.5 was just something else 😔😔 This will always be one of the best LLM launches
I will never forget Opus 5.5
1
24
1,122
T3code has changed the way I ship & do daily work. I was using it as a codex alternative since the codex app was terrible at that time. When the v2 sidebar dropped, it was outstanding, & after the orchestration v2, it's been a joyride. The team cooked hard on this; if you aren't using it, you are ngmi.
2
8
297
I normally don't do Multi agent work Everyday, but I use it every other day & it helps every much in that. There are many features there were released with orchestration V2 that I use & the Guided PR also. PR relations & the side bar is just the cherry on top & etc.
1
47
Had such a amazing time!! Got to meet @marclou !! Even tho wasn't able to talk cuz of time, but it was amazing! Thanks @Avi31m & @cognition team for this event & free creds!!!!
Amazing event with @marclou today, we had over 1.3k signups and such an amazing group of people at the venue today! was so happy how the community has grown since i came back to dubai to start this with @DevinAI / @cognition cant wait to keep bringing these events for you guys across the MENA region!!! Saudi soon....
6
26
626
Holy, no wonder my 5 hr was going faster than my weekly was used! Using Claude on T3code or through the Agent-SDK uses your 5-hour limit faster! Better to use it in Claude code until this is fixed!! Hope this gets fixed as soon as possible. Read this post by Theo 👇👇👇
Spent a lot of time on this today. I confirmed it's a real bug on Anthropic's side. To be clear: this does not affect weekly limits. All Agents SDK usage will drain your 5h limits faster than using Claude Code directly. I have reported this to Anthropic, hoping for a fix soon.
9
33
3,322