raymel 👋
@pseudokid
Followers
3K
Following
66K
Media
1K
Statuses
22K
AI tech guy for solopreneurs @02AIDEV | 3rd placer🏆 1st ever Vibe coding Web3 hackathon - Vibed 9 games in 3 weeks 🎮 | Want your AI built? Go → https://t.co/6CB8uRl2M1
MNL / Build your AI with us →
Joined September 2023
with the right people around you — being too busy becomes rewarding makes having "a lot in your plate" your own happy problem the secret? cultivating mutualism building on trust and goodwill is something you'll never regret — it makes reminiscing the journey rewarding
1
0
2
Introducing Lovable Connectors. Build more powerful apps by connecting to your favorite tools. Perplexity, ElevenLabs, Firecrawl, and Miro seamlessly connected with Lovable. Here's what this enables: (thread)
45
93
797
Agent is on another level In a few minutes, this run was able to output 240,000 data points It would have taken a whole team multiple weeks to gather this much information
Introducing /agent by Firecrawl 🪄 Just describe what you need - with or without a URL then /agent searches, navigates, and gathers information from the widest range of websites, reaching data no other API can. Try out the research preview today.
30
94
1K
Current LLMs support contexts with millions of tokens. However, we keep seeing failure modes due to poor long-context reasoning. Our new work shows that, for long contexts, we must perform test-time training updates rather than vanilla ICL or “thinking”! w/ @Meta & @KempnerInst
15
79
444
Congrats to @triggerdotdev on their $16M Series A! Prototyping AI is easy, but building production-grade agents is hard. https://t.co/hCeIIsz5dc handles reliability, scaling, orchestration, and observability—already powering hundreds of millions of agent executions per month for
15
20
167
"why would a company make something like this!?? what is the ROI!??" simple: i have seen this landing page now 3821 times it shows the craft and work they are putting into their work being a successful company means doing many things very well over longer periods of time
8
2
39
Whenever you see that I post less often, you know that I'm in build mode. 🤓 Honestly I'm not sure if it's good or bad. 😅 Whenever I'm building something interesting, I just don't have attention for anything else, including girlfriend, let alone social media. 😂
10
1
32
Harvard and Perplexity researchers conducted the first large‑scale field study of how people use AI agents. We analyzed hundreds of millions of anonymized user interactions in Comet to answer three questions: who’s using agents, how much, and what for. https://t.co/w9YrAKdvqz
perplexity.ai
Explore Perplexity's blog for articles, announcements, product updates, and tips to optimize your experience. Stay informed and make the most of Perplexity.
40
107
623
@karpathy we can finally call LLMs aliens now, lol. https://t.co/mYnggcYN3D
LLM training and inference in space. It's happening! This is why SpaceX is eyeing a 2026 IPO. This is why Google is sending TPUs into orbit. AI models will be born and will live in space. We can finally call them “aliens.”
5
4
39
i wish people understood both of these things 1. yes software quality is generally pretty bad and we need to figure out how to try harder 2. there is so much software that needs to be built - making it more accessible with higher level languages, frameworks, AI, services, is a
61
38
987
GLM-4.6V can read my horrendous hand writing and explain the math correctly Really loving this model, how well it does tool calling, how many languages it knows and its visual accuracy.
10
9
129
GLM-4.6V Series is here🚀 - GLM-4.6V (106B): flagship vision-language model with 128K context - GLM-4.6V-Flash (9B): ultra-fast, lightweight version for local and low-latency workloads First-ever native Function Calling in the GLM vision model family Weights:
117
257
2K
I’m changing my profile pic to a man Currently at 5031 followers. Let’s see where i’m sitting at the end of the year. Only the OGs will know.
I’m changing my profile pic to a girl Currently at 4833 followers. Let’s see where i’m sitting at the end of the year. Only the OGs will know.
223
27
3K
I’m changing my profile pic to a girl Currently at 4833 followers. Let’s see where i’m sitting at the end of the year. Only the OGs will know.
2K
339
44K
Big news: Replicate is joining Cloudflare. Replicate's going to carry on as a distinct brand, and all that will happen is that it’s going to get way better: it’ll be faster, we’ll have more resources, and it’ll integrate with the rest of Cloudflare’s Developer Platform.
264
170
2K
Antigravity should just start rolling out their pricing model at this point
Just when Claude releases Opus Google's Antigravity seems to come back to earth. Have had this for ±10 Minutes.
2
0
4
Just when Claude releases Opus Google's Antigravity seems to come back to earth. Have had this for ±10 Minutes.
Introducing Claude Opus 4.5: the best model in the world for coding, agents, and computer use. Opus 4.5 is a step forward in what AI systems can do, and a preview of larger changes to how work gets done.
1
1
7
It's a very difficult decision to wind down a profitable business But in this case, I think its the best one for current @gummy_search customers, as well as myself I wrote up the "why" here, as well as what what to expect in this final chapter https://t.co/LXxLJVWyXa
gummysearch.com
A message from Fed regarding upcoming changes to GummySearch
81
7
203
As a fun Saturday vibe code project and following up on this tweet earlier, I hacked up an **llm-council** web app. It looks exactly like ChatGPT except each user query is 1) dispatched to multiple models on your council using OpenRouter, e.g. currently: "openai/gpt-5.1",
I’m starting to get into a habit of reading everything (blogs, articles, book chapters,…) with LLMs. Usually pass 1 is manual, then pass 2 “explain/summarize”, pass 3 Q&A. I usually end up with a better/deeper understanding than if I moved on. Growing to among top use cases. On
901
1K
17K