record sessions w/ Claude Code, Codex, Gemini & Cursor into an indexed, searchable, and analyzed database for your agents. 🦞 https://xitter.cf/t.co/XcFjNlYcj9
Laying the Claw Law
Joined June 2025
- Tweets295
- Following135
- Followers188
- Likes444
tribecode 🔥 retweeted
General Catalyst just co-led a $31.5 million seed round into a blatant rip-off of my company, Kled.
(skip to 40 seconds if you want to skip context)
I would typically not speak on things like this, but this level of blatant copycatting is egregious and completely unacceptable, and needs to be made an example of.
This is one of hundreds of YC startups who have conducted this disgusting behavior. Unimaginative slop that continues to get rewarded due to nepotism.
Super excited to colead @LuelCompanyAI’s $31.2M seed round.
There are certain teams you meet where you know within 5 minutes that you want to partner with them Luel is one of those team.
William and Inigo are incredibly ambitious founders who understand the human data bottleneck from the inside out. They've built Luel to create a scalable, reliable supply of that data — something that will be foundational to the next generation of AI.
tribecode 🔥 retweeted
andrej karpathy got something right -
the renowned AI researcher today posted that "obsidian doesn't work with over 100 records" - and he's right
building a database out of markdown files is like building your entire filing system on sticky notes 🗒️
we benchmarked mitosis at 100,000 items
the trick is letting agents design the schema
you end up with the right connections, because it reinforces what claude gets wrong, instead of recording what humans think is important
get our guide on how to rag your whole life
comment 🤩 and i'll dm you the guide
tribecode 🔥 retweeted
Throwback thursday
We still have this lil guy in the office, but my actual clawdbot has been replaced with a hermes agent :')
tribecode 🔥 retweeted
ANDREJ KARPATHY COULD HAVE CHARGED $2,000 FOR THIS COURSE.
He put it on YouTube.
The full training stack. Tokenization. Neural network internals. Hallucinations. Tool use. Reinforcement learning. RLHF. DeepSeek. AlphaGo.
3 hours of the most comprehensive LLM education that exists anywhere at any price.
Not how to use the tools.
How the entire system was built from the ground up and why it behaves the way it does.
The engineers who understand this build things the ones who only use the tools cannot even conceive of.
The gap between those two groups is not 3 hours.
It is everything those 3 hours quietly unlock for the rest of your career.
tribecode 🔥 retweeted
Autonomous coding agent for relentless self-improvement
github.com/uditgoenka/autore…
tribecode 🔥 retweeted
This 1-hour podcast with Anthropic's Head of Product will teach you more on "how to make money with AI" than 100 articles in your feed
Bookmark it & give it 1 hour today. It will change the way you use AI forever. Then read the post below
Run a $50,000/month AI UGC content factory with Hermes Agent + Higgsfield Marketing Studio
in 2017 you copy a Shopify store, run a $20 ad → wake up to $800 in orders. This feels exactly like that.
brands paying $300-$800 per UGC video → you make it for 20$
start in 3 steps:
• drop a link to product / website
• create an AI persona - pick face / voice / vibe
• Hermes Agent handle the rest
no prompt engineering. No settings. Hermes scrapes, writes, designs, films, and posts. You just watch it run.
how it works in backend:
• hermes scrape {meta + tiktok ads} for best creatives
• seedance 2.0 - turns data into UGC ads
• AI UGC personas are spun up: realistic faces / voices
• agent work 24/7 producing 500+ viral-videos/day
grow the audience via Shorts/Reels, then sell AI UGC to e-comm brands at $300–$1k per video
static agencies are dead. AI UGC creators will print in 2025
tribecode 🔥 retweeted
We created a benchmark and found mistral considerably outperforms and costs less!!
Github.com/tribe-inc/claw-be…
We created a benchmark and found mistral considerably outperforms and costs less!!
Github.com/tribe-inc/claw-be…
Hard to believe there are no replies on this comment
And now its --dangerously-skip-purrrrrrmissions
Vibe coding is the future of development.
But we aren't going to 10x our progress by model improvements alone.
We need to turn inference into actionable insights.
Start tracking your agent conversations, and start seeing the difference.
tribecode 🔥
There's a new kind of coding I call "vibe coding", where you fully give in to the vibes, embrace exponentials, and forget that the code even exists. It's possible because the LLMs (e.g. Cursor Composer w Sonnet) are getting too good. Also I just talk to Composer with SuperWhisper so I barely even touch the keyboard. I ask for the dumbest things like "decrease the padding on the sidebar by half" because I'm too lazy to find it. I "Accept All" always, I don't read the diffs anymore. When I get error messages I just copy paste them in with no comment, usually that fixes it. The code grows beyond my usual comprehension, I'd have to really read through it for a while. Sometimes the LLMs can't fix a bug so I just work around it or ask for random changes until it goes away. It's not too bad for throwaway weekend projects, but still quite amusing. I'm building a project or webapp, but it's not really coding - I just see stuff, say stuff, run stuff, and copy paste stuff, and it mostly works.
Class is in session for clawdbotting on a budget with @cto_ya_know and Recursive Language Models (RLMs) with @neuralkian 🔥🦞 xitter.cf/i/broadcasts/1PlKQOMag…