Pinned Tweet
We have conducted our first workshop on how to fine tune a language model using @UnslothAI to be able to play the Codename board game at the University of Oklahoma! (@UofOklahoma)
Participants brought their laptops, signed up and logged in to their accounts, learned the game, played it without a language model as a player. Then inside their browser a gpu from our lab was automatically assiged to them, learned the training pipeline, trained their own model, and then played with their own trained bots as a player! All in 2 hours!
It was such a fun and unique event. Opus 5.5 was extremely helpful to make this event go smoothly by assigning everyone to an available gpu based on our capacity, and making sure everyone have their own jupyter server and also implemented such a nice UI to explain every single line of the training process.
We're only touching the tip of the ice berg what frontier models can do in designing interactive environments for education! Can't wait to have better events with interesting use cases of LLMs!
A huge thanks to our vice president @SaraAminpour to make this event possible!
I'm so happy for google deepmind for their amazing 4 Neon release :)
But I'm not sure why tf 4 Neon is saying "a very interesting session just appeared". even though I'm fixing a small bug in t3 code💀I guess frontier models love any kind of challenge
23
Arash retweeted
visualizing the transformer architecture in a nutshell
Link : poloclub.github.io/transform…
I think a lot of people aren't using Grok Bot to its full potential, because knowing how to automate things is actually quite a difficult skill to learn. I wrote this post to break down how I approach automating anything with Grok Bot!
Arash retweeted
you cannot automate what you do not understand
and to understand, you cannot first automate
slow down to go faster i guess
After using @VictorTaelin 's OptMem for 3 months, my agents know the context of my work and life really well. It's a game changer for people working with multi-agent environments. I also asked fable 5.2 to benchmark claude-mem (github.com/thedotmack/claude…) which has 118k stars on github and compare it with OptMem. The numbers speak for themselves :D
my AIE talk is up! come hear how the grok @bot team ships so quickly without compromise on quality
youtu.be/O_zcc6Xl12g
In early testing on FrontierCode 2.0, Fable 5.2 at High effort scored about level with Opus 5.5 at Max effort, for roughly half the cost per task. At Xhigh effort, it scored the highest of any released model tested.
Medium is the default in Claude apps and Claude Code.
It's straight-forward to connect grok bot with t3 code mcp server using tailscale, but how about dots from open ai?
For now you can't install tailscale on dots cloud computer (because of their network policies). "t3 connect" works, but it was not easy to setup.
So I made a skill so your dot can connect to t3 code mcp server faster:
github.com/Arash-san/t3-dire…
Why connecting t3 code to dots instead of grok bot? dots can have unlimited astra model calls, and to be honest, its voice engine is so much better than groks. It feels like freaking "her", the movie.
1
1
128
Arash retweeted
Thanks for being a great partner, Nvidia!
Congrats to @RicursiveAI on Ripple-65B Preview, their first open model 🎉
It was trained on NVIDIA B300 GPUs & sharpened with open NVIDIA Nemotron post-training datasets.
Download the NVFP4 checkpoint from Hugging Face and run locally on DGX Spark 👉 huggingface.co/ricursiveAI/R…
Arash retweeted
We will all be learning in coming weeks & months, that most human-made software:
▫️ Was really slow
▫️ Had many security vulnerabilities
▫️ Was under-specified and incorrect
AI will rewrite it all. Many times. It will make it fast, cost-efficient, and secure. It will think about failure modes we couldn't wrap our heads around.
It will prove, formally verify, load test, harden. It will turn memory-unsafe foundations into robust, optimized ones. If you ask, it'll thoroughly document it, teach you about it, make diagrams, animations, and slide decks.
In many ways, we are witnessing the true birth of software engineering.
maybe a hot take but i now consider adversarial reviews mostly a waste of tokens. far better to use those tokens on empirical combined with formal verification. i have updated pstack to use the interrogate skill less, and eventually i think i may remove it altogether
i have found very little value with code review in the abstract, agents tend to just get lost in the nitpicks and yagnis
Arash retweeted
i've reproduced the results from the MIT blog post today on weight-space programming. but one additional result i have to add: this is a new, step-change emerging capability that's appeared in just the latest set of models.
we will need to rebuild and rewrite much of software to adapt to a new, faster pace.
code used to arrive one drop at a time. soon, if not already, code will be written as if it were a continuous stream, like a violent river twisting and turning as it pleases.
this is not a new thing: big tech cos have had to deal with this with the enormous scale of their engineering teams. but now ai has brought an incredibly elastic and capable workforce wielded by every engineer. this is your problem now, and the only way to adapt is to evolve your systems so agents - even and especially the ones with little context about the right way of doing things - can thrive. make the easy thing the right thing.
your codebase is becoming organic architecture. agent contributions grow around the constraints and guardrails you’ve designed. good software engineering matters even more now! and no, the continuous stream of code is not something you have to watch helplessly. be more ambitious and design systems, frameworks, and languages to control the chaos.
the time for mourning coding is long over
Replying to @josephfcox
Here is Proton CTO's AI manifesto, that I obtained multiple copies of 404media.co/manifestos-fear-…
I asked Claude Opus 5.5 to explain today's new theoretical results from OpenAI. The Exponential Time Hypothesis being false is really crazy. What if the P vs NP problem is finally solved?
We are announcing three new discoveries in mathematics, computer science, and physics about harmonic analysis, complexity theory, and quantum field theory.
openai.com/index/building-sc…
1
1
425
































