If you want to start a startup, don't learn "entrepreneurship." Learn how to build things. The hard part of startups is not "entrepreneurship" but product: to know what to build, and to be able to build it.
If a user asks "What's the capital of France?" and another asks "Tell me the capital of France", the second query hits the local cache. Reduces LLM latency & API cost.
Built for performance in Rust using Axum and ONNX.
https://t.co/3ew5dxWx8s
I built Semantic Cache Gateway that sits between your users and the LLM. It generates vector embeddings of incoming prompts locally and uses a Qdrant database to detect semantically identical queries.
built semantic gateway in rust, which reduces your LLM token usage by filtering out same question of around 0.95 cosine, more detailed post and github repo incoming before midnight.
the difference between my ambitions and the level of work i do is insane, i need to get my level and amount of work close to the level of ambitions i have, i can't lower the ambition but i can sure increase the work amount
The richest people I know go from idea to action immediately. There’s no analysing. It’s just straight implementation. You’d be much further in life if you acted as soon as you thought. Try it.
networking as activity is mostly cope.
e.g. the conference circuit, the warm intros, the moving to sf discussions or whatever, oh & the “grabbing coffee” economy.. all of this is overwhelmingly negative selection esp with vc (lol). the ppl worth knowing are usually too busy doing the thing to be farmable, & the ppl available to be networked w/ are available cuz they have literally nothing better going on.
do the work, then publish it loudly enough that the right ppl can find you w/o you having to chase. one way broadcast > two way schmoozing. this is why x matters a ton now more than ever before.
The 2-4 hours you spend scrolling each day (or 730-1460 hours each year) is more than enough time to write a book, build a business, or get in shape. In the moment, it seems like nothing. That's why it's so dangerous. Your time disappears without you being conscious of it.
> wrote an inference optimizer which reduced my ram footprint from 2gb -> ~700mb (~62.7%) and latency from 2.34s -> 1.36s (~42.6%)
> ran on my i7 Iris Xe iGPU of i7-1185g7
> written in c++ using intel openvino toolkit
> squeezed 32bit floating model to 8 bit integer model