Joined October 2012
- Tweets 539
- Following 319
- Followers 1,645
- Likes 5,597
31 Photos and videos
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
222
233
2,524
415,685
with the events around fable, it’s clear that companies & developers need to own their models.
the ability to post-train & rl models must become a more broadly accessible skill.
rl fine tuning sounds like rocket science, but it really isn’t.
so @Thariq_q (not the claude code guy :D) made a video that explains it with as little jargon as possible👇
5
5
25
1,408
💯 agree. we are building @castformai to empower any developer in the world to be able to do this
When you rent your artificial intelligence, you have no control, and no choice. This is why sovereignty and ownership matters.
Whether it means using your own hardware, open source, or deep customization. Own your AI, own your future.
2
2
14
1,668
one of the most exciting surprises from the launch has been seeing just how diverse the developer use cases are for custom models. human creativity is really unmatched :)
we really want to help bring more of these ideas to life.
so, we’re giving away $1,000 in free credits to folks with compelling ideas they want to train.
just reply to the post with your idea, or email us at castie[at]castform[dot]com
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
14
9
69
14,463
girish retweeted
Jun 12
This looks like a good way to get started with training your own model.
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
3
5
1,098
girish retweeted
Jun 11
Quite unique and interesting startup. Launch video is really good and elaborative the website is also interesting.
Check it: castform.com
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
1
3
9
1,443
highest praise i have gotten in a decade 😂
Jun 11
incredible launch video
- very very original concept
- founder is super authentic and compelling when he talks
- elite editing given how few different shots there are
- animation on the character is spectacular. looks like a pro worked on this.
- dialog between founder and character is great and paced nicely
- brilliant idea focusing on key questions people ask about their space
- sound fx and music is solid
- great scroll stopper, bold statement, social proof, etc
but
- only feedback is to add captions
this is might be my favorite launch video of last 6 months.
the biggest thing they nail is simply copying a PROVEN format from YouTube/film/TV.
the physical board with google queries on it is something I've seen on tons of YT videos but have never seen on a launch video. and the animated character closely parallels pixar's characters.
had they tried to change even a small piece of this or not executed it well, this wouldn't have hit as hard.
not only is this an easy hack for coming up with video ideas, but it ALWAYS works. Founders Fund just did it with reality TV, TBPN did it with ESPN.
spectacular work
3
26
4,532
every company will have a custom llm like every company has a website :)
Jun 11
seems likely that this time next year there will be fewer ai labs (as pretraining keeps scaling) but lots more people post training. congrats girish!
4
3
23
8,451
girish retweeted
Great to see the launch @castformai! Long past due for a simple devtool that enables customization and post training
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
5
21
3,972
girish retweeted
Jun 11
castform is in open beta!
our goal’s to enable any developer post-train their own llms. in the world of rapidly rising llm costs & providers guarding capabilities, we believe the ability to shape model behavior shouldn’t be a privilege.
this release today is our small step towards fixing that.
$50 in free credit for new users. hope you’ll give it a try 🙂
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
6
6
43
7,672
“don’t train your own model” is common ai advice. it's wrong. your token bill's the proof.
today, we’re excited to launch castform into open preview. castform is the easiest way for you to train your own model, on your own data.
open-weights models are performant and much cheaper. when trained on your task & proprietary data, they beat closed models. the thing standing between you and that was weeks of plumbing & years of ml expertise.
with castform, model training is as simple as prompt engineering. @castformai
bring your agent traces or raw corpora. castform turns it into training data, picks the right algorithmic recipes, manages gpus, and gives you an ide to watch and chat with your model as it learns.
see what you can build with castform👇
222
233
2,524
415,685
train your first model today at app.castform.com
castform — Training Platform
train, evaluate, and ship custom models with castform.
app.castform.com 2
28
3,763
AI is creating problems it still can’t solve.
The same technology poised to automate millions of jobs still can’t reliably help people navigate SNAP — the food assistance program 40 million Americans depend on. We built the first benchmark to measure that.
Partnering with Center for Civic Futures and @codeforamerica , we scored models on SNAP question scenarios which users would have to navigate, with expected response rubrics validated by policy experts to match practice considerations.
The best model only scored 62%. Models handle federal questions like appeals and recertification reasonably well, but fall short on state-specific ones like replacing an EBT card. Benefits are administered by the states, meaning models are weakest where people need them most.
The same technology poised to automate millions of jobs should at least help strengthen the social safety net for the people it could displace. As AI adoption proliferates into public services, governments need a reliable way to test these tools before deploying them.
14
15
104
16,821
What happens when you let an llm attack itself on repeat?
Attacker finds jailbreaks → those become defender training data → repeat. Defense rate went 64% → 92%, no human-written adversarial prompts.
we trained Qwen3.5-4B with RL to get itself to comply with requests about making meth and stealing credit cards.
then we used the attacks that worked to train the model’s defenses, and repeated the loop - fully automated red-teaming.
defense rate went from 64% → 92%.
1
8
664