sarusso

For context: a LOT of people, maybe naively, are now using AI to help them count carbs, and some of these features are already in beta, if not shipping.

That is why I believe this piece from Tim is remarkable: it shows the limitations in a language the diabetes community can understand, and this is why I posted it.

The "specific set of actions" is so vague that could range from just opening a specific company page and clicking on a button to performing a complex chain of steps.

This said, it's not that bad, that's true. But the idea of having the personal residential address exposed is not great either.

"Our investigation found that it was technically possible for a logged-in registered user to:

1. See certain data not normally published on the public register:

- the day of the date of birth for directors and PSCs

- residential address for directors and PSCs

- company registered email address

2. File updates to any information without consent. For example, new accounts or changes of director."

Location: TBD (Italy for now)

Remote: Yes

Willing to relocate: Yes

Technologies: Python, Django, ML/DL, HPC, Docker, Kubernetes, CI/CD, observability, time series modeling, anomaly detection

Résumé/CV: https://sarusso.github.io/CV-Stefano-Alberto-Russo.pdf

Email: stefano dot russo at gmail

I'm a tech lead with interdisciplinary experience across research, industry, and entrepreneurship. I’ve worked at CERN, collaborated with major industry players, and joined startups and scaleups. I also co-founded a startup and participated in Entrepreneurs First.

Now looking for my next challenge, feel free to reach out!

What I did in a few projects to plot aggregated (resampled) data without loosing peaks was to plot it over an area chart representing the min-max values before aggregating (resampling). It worked pretty well.

Happy to hear! It doesn't replace healthy bedtime routines of course, and it's halfway between an experiment and a joke, but still... :)

That's interesting. How would you explain that for someone works then? At least on me, the loader does not trigger vigilance and temporal monitoring (I think). Instead, it "allows" my brain to focus on something else, which gets boring very soon, but that at the same is engaging enough to keep doing it. This kind of interactive yet very slow dynamic helps me stop trains of thoughts and relax.

It does not replace healthy bedtime routines of course, and it never meant to be a serious sleeping aid, but more of an experiment - and partially a joke. Maybe the premise should have been to help stop a spinning mind rather than to fall asleep...

The skeleton provided by ChatGPT to get the page to load fullscreen as a PWA included it, so I just kept it. “Maybe in future”... But it’s unnecessary, I should have removed it.

Feel free to send a PR! But let's parametrize the animal :)

Exactly, I made a mistake when writing it down. Thanks to the original commenter for catching it and for your suggestion!

I’ve now updated it to: "As you go through the story, the time the loader spins increases while the speed at which the text appears decreases"

Oh yes, absolutely. It bothers me as well, and it strains the eyes, but you know... for such a toy project I half vibe-coded in an hour or so I didn’t bother too much. I agree it should be improved!

It could definitely be a fun exercise. Also maybe just rendering all the text in the same color as the background and then changing the colors of the characters one by one could be an interesting option (just thought about it), but I think yours would render better.

As a side note, I have to say that posting something as simple as this, where you can’t really get too attached to the project and can read feedback in a truly neutral way - instead of just pretending - is so refreshing...

Thanks for your comment!

I came up with this one night when I couldn’t fall asleep and was looking (waiting) for some data analysis results on a web platform, noticing that the spinner was actually making me sleepy.

I guess that when the brain is engaged in an activity, and that activity becomes boring, it creates good conditions for falling asleep.

  Location: TBD (Italy for now)
  Remote: Yes
  Willing to relocate: Yes
  Technologies: Python, Django, ML/DL stacks, HPC, Docker, Kubernetes, Slurm, and whatever I needed to build data&resource-intensive systems
  Résumé/CV: https://sarusso.github.io/CV-Stefano-Alberto-Russo.pdf
  Email: stefano dot russo at google mail

I’ve worked at CERN, in startups and scaleups, and at international research centres. I’ve also co-founded a startup and participated in Entrepreneurs First. I’ve led teams and owned technical and product roadmaps end to end.

I’m primarily interested in lead roles and above (Staff/Principal/Lead/Head), including founding or early technical leadership roles, with substantial ownership and high agency, where I can define strategy and execute it. I’m particularly effective in interdisciplinary, cross-field, and ambiguous environments, where I can leverage my zig-zag career path to act as a force multiplier.

I’m also open to senior IC roles in areas where I’m particularly strong and can be a “10x”: time series, anomaly detection, and making HPC easier to use.

If you have anything for me, I look forward to hearing from you!

Just found a strange switch on a CD burning app (Burn-OSX) which stated: "Pre-emphasis of 50/15μs". Turned out to be a historical hidden gem, aimed at "boosting the signal (especially the higher frequencies) in the recording before it was put onto CD, and embedding flags in the disc's subcode to tell a CD player to apply de-emphasis on playback. Some CD players even had a de-emphasis button that could be used to manually apply de-emphasis".

Thanks!

This example is so right. Including the parallel with what happened with those two aircrafts.

The communication between your phone/pump or glucose sensor/pump is encrypted now for all newer devices.

May I ask where did you get this info? And what “newer” means here?

Great to hear! p.s. I just copy-pasted to help, not to critique.

Remote (US-based or Canada-based only). Must be authorized to work in the US (no visa sponsorship available).

The main flaw of this article is comparing a general-purpose language built with production systems in mind (Python) with a domain-specific language designed for interactive analysis (R)... Beware of comparing apples and oranges, because productizing R code typically requires rewriting it in another language.

Do they run but look terrible or they just refuse to open?

"They result in deliverables"

I had a _so_ different experience after being involved in two of them...

On what basis are you are stating this? I'm asking because I have been involved in another project like these (15M budget) and the main issue was the lack of computing resources allocation, because no one thought about it (true story).

DeepSeek had plenty of R&D expertise which were not included in the (declared) model training cost. Here we are talking about building something nearly from scratch, even if there is an open source starting point you still need the infrastructure, expertise and people to make it work, which with that budget are going to be hard to secure. Moreover these projects take months and months to get approved, meaning that this one was conceived long before DeepSeek, thus highlighting the original disalignment between the goal and the budget. DeepSeek might have changed the scenario (I hope so) but it would be just a lucky ex-post event… not a conscious choice behind that budget.

They allocated €37.4 million [1]. As an European, I truly don’t understand why they keep ignoring that the money required for such projects is at least an order of magnitude more.

[1] https://digital-strategy.ec.europa.eu/en/news/pioneering-ai-...

Location: Europe/Italy

Remote: Yes

Willing to relocate: Depends on the opportunity

Technologies: Python, Django, Docker, ML, time series, IoT, anomaly detection.

Résumé/CV: https://sarusso.github.io/cv.pdf

Email: stefanorusso+yc@gmail.com

I am a hybrid profile, somewhere between a software engineer, a scientist, and an entrepreneur. At heart, I am a builder who loves solving problems and making things work. Right now, I am on a non-linear career path, and still waiting for the right opportunity to fully commit again. My website: https://sarusso.github.io/

I recently built a small Python library to try getting time management right [1]. Exactly because of the first part of your comment, I concluded that the only way to apply a time delta in "calendar" units is to provide the starting point. It was fun developing variable-length time spans :) I however did not address leap seconds.

You are very right that future calendar arithmetic is undefined. I guess that the only viable approach is to assume that it works based on what we know today, and to treat future changes as unpredictable events (as if earth would slow its rotation). Otherwise, we should just stop using calendar arithmetic, but in many fields this is just unfeasible...

[1] https://github.com/sarusso/Propertime

I remember hearing at a conference about 10 years ago that Google does not make use of leap seconds. Instead, they spread them across regular seconds (they modified their NTP servers). I quickly searched online and found the original article [1].

[1] https://googleblog.blogspot.com/2011/09/time-technology-and-...

We are already discussing this here: https://news.ycombinator.com/item?id=40477166#40481115. No need to cross-discuss.

I actually went into every single line of the manuals and even discussed the matter on the official Singularity Slack.

In that blog post I wrote that it does not enforce. It is true that you can achieve some level of isolation by setting certain flags and environment variables explicitly, but this is (was?) quite hard to get working, moreover the user mapping inside the container is always host-dependant and there is just no network isolation.

To achieve something close to the behaviour "I wanted", I had to use a combination of the command line flags you mentioned (and in particular -cleanenv, -containall and -pid) together with custom-made, ad-hoc runtime sandboxing for directories which required write access (as /tmp and /home).

However, this is not the default behaviour and it is not how Singularity is used in practice by its users. But yes, I was able to achieve something close to the behaviour I wanted [1].

This said, if I am missing something, or if the project has evolved to allow for a better level of isolation by default, please let me know. That blog post is dated 2022 after all.

[1] https://www.sciencedirect.com/science/article/pii/S221313372...