korrents
Salvatore Sanfilippo

Salvatore Sanfilippo

@salvatore-sanfilippo · 20 positions · On ppll ↗

Programmer; wrote Redis and hping, and blogs about C, systems programming and working alone.

Salvatore Sanfilippo did not write this page.

We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.

  1. AI alignment

    the first serious AI incident is very likely to happen *inside* the walls of frontier AI labs, while testing a new model, or while the AI lab employees, or the few externals who have access, do something wrong compared to the expected power of the model.

    ↗The real AI risk is inside the labsantirez.com 1st of 3 in this piece

  2. LLMsopen sourceAI alignment

    The real risk is leaks, not releases, and leaks happen inside frontier companies.

    ↗The real AI risk is inside the labsantirez.com 2nd of 3 in this piece

  3. AI alignment

    a single company should not be able to evaluate independently whether a model is safe enough. We need a joint AI safety organization that includes experts from all over the world and is recognized by the governments where frontier AI companies exist.

    ↗The real AI risk is inside the labsantirez.com 3rd of 3 in this piece

  4. 3 days earlier
  5. ambition

    Writing a kernel is not something within everybody's reach, but it is something that many can complete, if they put enough effort into it.

    ↗Being Linux Torvaldsantirez.com 1st of 3 in this piece

  6. vibe coding

    Vibe coding is a wrong idea of what automatic programming is, and of what automatic programming will be for the majority of people.

    ↗Being Linux Torvaldsantirez.com 2nd of 3 in this piece

  7. LLMscoding agentsdesign

    Automatic programming, instead, in the hands of people that are expert technicians, or expert programmers, expert designers, expert software architects, is to assume the role of Linus, with the agents and the LLMs assuming the role of the different maintainers of the different subsystems.

    ↗Being Linux Torvaldsantirez.com 3rd of 3 in this piece

  8. 3 days earlier
  9. design

    Because of this change, the idea of just having a stable branch with everything polished, and an unstable branch where everything is a work in progress, may no longer be the right way to do things.

    ↗Not just development, distribution of software may change as wellantirez.com 1st of 2 in this piece

  10. coding agentsdocumentationdeveloper tools

    Also, it means that the documentation itself should not be just good for humans, but also for coding agents to understand how to change the system.

    ↗Not just development, distribution of software may change as wellantirez.com 2nd of 2 in this piece

  11. 9 days earlier
  12. ambition

    I believe many programmers at this point have less impact they could have because they look at the code.

    ↗Control the ideas, not the codeantirez.com 1st of 3 in this piece

  13. vibe coding

    if you control the ideas of your software, looking at the code itself is suboptimal and often pointless.

    ↗Control the ideas, not the codeantirez.com 2nd of 3 in this piece

  14. LLMs

    LLMs are very good at writing locally optimal code, and are worse (but improving) with big ideas.

    ↗Control the ideas, not the codeantirez.com 3rd of 3 in this piece

  15. 5 weeks earlier
  16. code generation

    In my experience the output does not reach the structural quality and economy of complexity of the best hand-written software.

    ↗A new era for software testingantirez.com 1st of 3 in this piece

  17. software quality

    It is a known fact that covering all the lines of the code does not mean covering all the possible states.

    ↗A new era for software testingantirez.com 2nd of 3 in this piece

  18. LLMscode generationsoftware quality

    I have the feeling that the introduction of automatic QA may raise the bar of quality for new releases of software, and maybe partially compensate for the lower quality of the code produced at high speed with the use of automatic programming.

    ↗A new era for software testingantirez.com 3rd of 3 in this piece

  19. 13 days earlier
  20. data centerssemiconductors

    There is almost zero hope that NVIDIA setups will get less expensive, and even a small company can't afford to easily purchase and handle a small data center for local inference.

    ↗Distributing LLM inference in DwarfStarantirez.com 1st of 2 in this piece

  21. LLMsopen source

    Open weights models are now in a golden age, we have plenty and many are very powerful.

    ↗Distributing LLM inference in DwarfStarantirez.com 2nd of 2 in this piece

  22. 6 days earlier
  23. LLMscoding agents

    This means, basically, that just using line numbers is very fragile: to say, change line 22 with new="foobar" is not good.

    ↗Alternatives for the EDIT tool of LLM agentsantirez.com

  24. 5 days earlier
  25. LLMssimplicity

    It is clear that there was a need for single-model integration focused local AI experience

    ↗A few words on DS4antirez.com 1st of 3 in this piece

  26. LLMsopen sourcesoftware performance

    The space will be occupied, in my vision, by the best current open weights model that is *practically fast* on a high end Mac or "GPU in a box" gear (like the DGX Spark and other similar setups).

    ↗A few words on DS4antirez.com 2nd of 3 in this piece

  27. the cloudSaaS

    it was really appreciated :) AI is too critical to be just a provided service.

    ↗A few words on DS4antirez.com 3rd of 3 in this piece