Linus von Ekensteen Löfgren

co-founder @halmoslabs | YC F26

[dax @thdxr]

a weird inversion with LLMs is the models improve faster than the tinkerers

when i see people with custom workflows and setups they're all addressing problems that don't exist anymore

the person naively using vanilla codex is more likely to be experiencing state of the art

[Linus von Ekensteen Löfgren @linusvekensteen]

Very bullish on more abstractions throughout the stack that makes good choices the default

[Garry Tan @garrytan]

AGI Science Loops are coming, and there will be Muse/Instinct for those AGI Science Loops. Halmos is building that.

[Linus von Ekensteen Löfgren @linusvekensteen]

@kul @halmoslabs @MattiasAkke @ycombinator Thanks Kulveer!

[Linus von Ekensteen Löfgren @linusvekensteen]

We started Halmos Labs because we are convinced that research is about to change, and we want that change to be for the better.
Through validation and verifiably we elevate researchers, so we don’t turn humans into just a proxy between machines and the world

[Halmos Labs @halmoslabs]

The world spends roughly $100 billion a year on experimentation in Life-Science. Yet, 93% of drug programs entering Phase I never reach approval.

Halmos Labs is changing that

See how at https://t.co/iEVkG2oWfj

[Linus von Ekensteen Löfgren @linusvekensteen]

Domain driven design at every layer of your stack is the one true way

[Matt Pocock @mattpocockuk]

My hot take from chatting to @poteto is that we should use MORE abstractions in the AI age

You can use them (combined with harsh lint rules) to reduce the design space available to the agent and constrain them only to good decisions.

Combined with the fact that high-leverage abstractions let you do more with less code - so, more token efficient.

Plus, unwinding the damage from a bad abstraction is much cheaper with agents.

This runs counter to a lot of folks thinking that agents just want to read the raw code. They can, but they're not maximally efficient that way.

Be braver! Design abstractions.

[Anshul Kundaje @anshulkundaje]

AI folks who are new to bio should spend some time with wet lab folks not only to understand how experiments, assays, lab work is done but also to learn the basic concept of CONTROLS. 1/

[Linus von Ekensteen Löfgren @linusvekensteen]

”You do it because it’s the right experiment to run, and you can’t cut corners […]. Running the mediocre experiment because it's easier or more feasible doesn't get the data you actually need” is exactly why intelligent exp design that elevates your team is so important

[Ron Alfa @Ronalfa]

you still have to do the hard stuff: to build an ai company in bio, you also need to build a bio company.

https://t.co/vf9y1IH2LH

[Linus von Ekensteen Löfgren @linusvekensteen]

proper DDD will save you from so many mermaid diagrams

[wits @witsdev]

we do this with overscroll-behavior: none

but we gate it behind a (pointer: fine) media query, so that it doesn’t impact touch devices. this means you can still pull to refresh on mobile