Congrats on the launch. We are using it for our emails out to folks and it’s working well for us. I also had the pleasure of working with Adam for a while when I hired him to join the Training group at Datadog soon after I started the team. When he said he was leaving to start an email company I thought he was crazy. But it’s a really cool product.
But I think we are there for most tasks. I thought a bit about this today and wrote a blog post about how to use a local ai tool, Ollama, to summarize, make links, and ask questions about my notes. It’s not a fully fleshed out plugin but shows the start in about 20 or so lines of typescript.
I am using ollama today on a MacBook Pro M1Max with 64GB. Using a llama2 70b model, I am getting about 7 tokens/second with the onboard gpu. Before ollama used gpu, that was much slower. To compare, the 7b model gets me closer to 55 tokens/second. There is no way it could achieve those numbers without the gpu.
Wow, amazing that this is still being developed with a lot of competition for a lot of that time. Curio first was released nearly 20 years ago. Devonthink started in the late 90s. and Tinderbox started in 2002. And each have huge numbers of customers that have stuck with them for decades.
That’s kind of the only way I have seen them. Individual teams define the okr and they bubble up. I don’t know what summarizes as their own would mean, but the first part was right.