On an induvidual level people are beating them by creating vapourware Ai startups and selling them to the big tech companies for stupid valuations, before proceeding to the roof for a bit of rest and vest.
From the lastmile team here. Would love your feedback!
Origin story -- we're a group of ML developers who want to build better developer tools to make AI more accessible to non-ML engineers. We built ML Ops tools in Big Tech for many years (Jupyter notebooks, model management, experimentation platforms, etc), but those were targeted towards ML researchers.
With LLMs and other generative AI models, we want to provide simpler tools to bridge that 'last mile' gap between the core foundation models, and using/personalizing/integrating them in your applications. Hence the name :)
Yes, many, huggingface is full of chat-tuned LLaMA derivatives that are supposed to replicate its performance, and tools like text-generation-webui or kobold.cpp can be used to run them with chat-style UX.
But for most tasks none of them come within a mile of GPT-3.5, or within a parsec of GPT-4.
Very well written and so true.
For myself, I have trouble dealing with people who want (need?) to prove they're here. Most of the time, they are overdoing it, and it is clear that it's about themselves.
Just see if people are responsive to your "support". Otherwise, You're just making it worse.
In some of the companies I've worked at, design was the last priority and at some point, making the effort to switch to a proper design system and do the front-end migration seemed too daunting. So it was never prioritized, until the exit.
I doubt there's a formal answer to this, but to me it's like 'branding': won't give immediate results but compounds over time.
And if you're a company who's never done branding, catching up after 3y is possible but super hard.
At the end of the day, I think it depends on the founders' sensibilities, because with no measurable ROI there are all the reasons in the world to deprioritize that!
I think it's never too late to implement a design system, interesting fact is the more you're waiting the more work you'll have to do 'cause you'll have to deal with more migration. At least one way to tackle it is to start with the foundation like a visual language systemised.
We all need such tactics rn!
Billing is still quite a pain, especially usage-based, when you're growing fast and iterating on your pricing.
Another significant pain is cash collection: it takes too much time for poor results and antagonizes customers. + there's a compounding effect of your uncollected invoices every month to the point where you end up writing off invoices...
At Upflow (YC W20), we consider that MRR should be what you collect, not what you invoice.
That's why we help companies collect their cash faster with our 2 click integrations to your invoicing tool and customer payment portals.
YC companies can actually benefit from a great discount right now => look for it in Bookface.
EUR feels still stuck in a world where you have to figure out everything from the get-go. Hence, I think Europe still attaches a stigma to pivoting, whereas it's a rite of passage in the US indeed. Hopefully that kind of posts helps changing that!
So true! I’ve met several companies living this exact scenario on billing: trying to balance building vs buying. None being acceptable solutions.
The first one dedicates so much of your developers on a project that a actually never ends. Which is obviously not an viable option.
The second is so expensive you end-up considering the first… and most of the time you still have to fit a circle in a square.
