Brev: Start fine-tuning and training models in < 10 minutes | Hacker News Reader

Brev: Start fine-tuning and training models in < 10 minutes

brev.dev

48 pointsomarfarooq7 comments

I am shocked at how this start-up with a not-great name that was just another developer environment tool not less than six months ago managed to do a full re-haul and pivot toward fine-tuned AI models. It's much more compelling now.

Popped into my feed recently with two laughing attractive employees. That little video loop of them popped back into my feed again, and again, and again. And then they did a new, expanded version. It's TikTok marketing crack.

They caught lightning in a bottle in a sea of high interest rates and startups going bust left and right. Fair play. Great combined maneuver from marketing and product.

Hey! Co-founder of Brev.dev here and was in that video. We're actually a small team of 3-- makes it easy to maneuver when marketing and product are the same :)

We were building cloud dev environments in 2021, which I now feel strongly do not have PMF. Writing a fun post on that...

We were trying to make coding in the cloud as seamless as coding locally. Users started using that for GPUs ~1 year ago. A solid dev env experience makes sense for AI/ML developers. The infrastructure problems are more complicated & more expensive.

We've since reworked our product and aim to build AWS SageMaker but without the confines of any particular cloud. We're starting by sourcing cheap GPUs & having 1st class notebook experience for fine-tuning and training. Inference coming soon!

and thats not even ALL the laughing attractive employees that they have!

brev's execution to make LLM finetuning approachable, educational, and fun has been nothing shy of inspiring

How do you source GPUs given the current GPU shortage? Parntership w Coreweave?

We work with smaller data centers that have single node GPU capacity. So while there's a shortage for clouds that have capacity for clusters, we're focusing on aggregating single nodes behind a unified API. It works for our users too-- a single A100 is more than enough to fine-tune llama2 or Mistral7b on your own data.

Also, as you can imagine, the smaller data centers have a pretty rough experience for provisioning. Our API handles difficulties from those smaller data centers to make the experience consistent with provisioning from a major cloud service provider.

I love Brev

Reply on news.ycombinator.com