lytefm

I've been working in this space for seven years, this isn't new. There have been big studies from the UK in 2020 [1] and Israel in 2015 [2] about this.

Some of my learnings:

- Don't start your day with a large amount carbs. Have some insulin in your blood before eating that big bowl of oatmeal. Or just go for some yougurt with nuts and seeds.

- The classical order of a three course meal (salad first, then main dish, then dessert) is pretty good in terms of preventing glucose spikes.

- Going for a walk after a meal is great for bringing glucose levels down.

- Eat at least 2h before going to sleep. Having high glucose levels disrupts sleep.

- Alcohol lowers the glucose response of a meal, but is still bad unfortunately.

- Diet Coke works. No spike vs loads of sugar with a real coke.

- Stress can spike glucose like crazy, e.g. being in an interview or during takeoff.

- If you really want to know how you react to some food, keep the circumstances (time of day, sleep, physical activity, stress) similar. There's too much influence beyond just the meal.

I bet that everyone who is wearing a premium smartwatch or an Oura ring now will be using a CGM now and then in the next years.

[1]: https://pubmed.ncbi.nlm.nih.gov/32528151/ [2]: https://pubmed.ncbi.nlm.nih.gov/26590418/

Yeah they bought veri.co last year and announced the integration of metabolic into their platform last week.

As the authors mention in the end, personalized nutrition based on CGM only makes less sense than also integrating sleep, stress, movement,... - which Oura is really good at.

> I think we will still need software devs, but not as many as we do today.

I'm more of an optimist in that regard. Yes, if you're looking at a very specific feature set/product that needs to be maintained/develop, you'll need less devs for that.

But we're going to see the Jevons Paradox with AI generated code, just as we've seen that in the field of web development where few people are writing raw HTML anymore.

It's going to be fun when nontechnical people who'd maybe know a bit of excel start vibe coding a large amount of software, some of which will succeed and require maintenance. This maintenance might not involve a lot of direct coding either, but a good understanding of how software actually works.

I'm not surprised. Apple ID seems completely broken to me, even when I just tried to set up a new MacBook 4 weeks ago:

Create an account via the settings page? Had to contact support because my phone number + email got stuck in some "halfway activated but can't log in" state.

Sign into the App Store + downloading some software? Got stuck in some broken form that wouldn't let me enter payment information. Had to crawl through reddit threads about how to work around this mess, just to be able to use the App Store.

I'd be fine calling myself a "AI Engineer" for having worked with TensorFlow, PyTorch and scikit-learn and having experience in running AI Models in production.

Nowadays, making requests to the API of a LLM provider is what makes an engineer a good fit for working at an AI company? Really?

Cold emailing someone who makes their contact information publicly available and might be interested in a sales pitch is not illegal in Europe. Sending SPAM is. The lines get even more blurry with automated AI tools that offer personalized sales pitches as a service.

I'd be curious how this plays out in court. Probably something like:

- If you use an AI tool to scrape leads and to generate the content but then still send out individual emails from your Mail provider, it's still a cold email.

- If you use an AI tool and also automate the email delivery, it should be considered spam.

> Does this mean that I should private my GitHub-mirror to my personal blog, because this can become a common thing? I have removed my email from my GitHub-profile now, but they can probably get it from my Git-log anyway...

It's possible to use a noreply.github.com linked to your username for making commits. And you can to change the authorship of past commits in your own repos with write access.

I try to avoid give my email in a public and processable format whenever possible.

This article reads as if the author was exploring the "Gervais Principle" [1] with Sam Altman as example.

> Of all organization men, the true executive is the one who remains most suspicious of The Organization. If there is one thing that characterizes him, it is a fierce desire to control his own destiny and, deep down, he resents yielding that control to The Organization, no matter how velvety its grip… he wants to dominate, not be dominated…

[1] https://www.ribbonfarm.com/2009/10/07/the-gervais-principle-...

Early this year, I've received a hostile PR for a "maintenance only" JavaScript authentication library with less than 100 stars but which is actively used by my employer.

It added a "kinda useful but not really needed" feature and removed an unrelated line of code, thereby introducing a minor security vulnerability.

My suspicion is that these low quality PRs are similar to the intentional typos in spam emails: Identify projects/ maintainers who are sloppy/ gullible enough and start getting a foot in the door.

Apples Stock price is has pretty much remained flat YTD whereas Google, Meta and Microsoft have ballooned thanks to the AI hype.

> The executive team is saying "the best thing we can think of to spend this money on is shrinking the company". That's not a good sign, even if they are also spending big on R&D.

As an executive, buying back shares makes a lot of sense if you believe that your company is currently undervalued and you have a large cash pile.

Tim Cook seems to be very optimistic about Apple's future [1], so doing a buyback in the current market might be a smart move.

[1] https://finance.yahoo.com/news/apple-ceo-tim-cook-boasts-of-...

> All of this has caused me to spiral into depression the last 1.5 years. Is it possible to get out of this situation?

Honestly, I'd doubt that working with legacy code simply causes you to feel depressed.

You might have an actual depression and changing jobs might just make things worse.

Go see some therapist, at least do some online screening for symptoms of depression.

I've had a very good time at my first job where my only task was to maintain a horrible legacy Java codebase because I wasn't depressed.

I've had some tough months in my current job at a startup, working with modern NodeJS, Python + ML, Docker stack because I was depressed.

Have you tried out working on multiple projects or tasks at the same time?

When I hit a wall, it's often futile to continue working on that single problem. So I'll just leave it as is is for a day and work on something else.

When I revisit task one the next day, I often have a fresh approach or I'm able to spot that bug I just couldn't find the previous day.

Other devs seem to prefer working very "linearly", but I'm doing better if I juggle several tasks at once.

I have the impression that some people can be software developers for 10 years or longer but they still don't "get it" while some junior devs are really great right from the start.

These Seniors might be good at duct taping, overengineering or have a "my way or the highway" mindset but aren't capable of finding and implementing a good solution given the constraints.

It probably boils down to factors like intelligence, creative problem solving, ability to focus, reading skills, knowledge retention and a healthy dose of humility.

I guess everyone can (and probably should) learn basic coding skills, but it doesn't mean everyone can become a great software engineer.

This could even play out on a broader scale and even increase the demand for software engineers.

What is going to happen if more and more people are creating lots of software that delivers value but is difficult to maintain and extend?

First it's going to be more high-level stuff, then more plumbing, more debugging and stitching together half-baked solutions than ever before.

AI might make our jobs suck more, but it's not going to replace them.

Yeah if it's more like "you work for me; I pay you as long as the results are good but will fire you at will" it's much easier to pull this off while maintaining a clear conscience.

> As long as they perform the tasks of their jobs, why shouldn't they be "getting away with it"?

I'd say it depends a lot on the legal framework of employment contracts. I don't know how it is in the US, but in Germany you sign a contract with your employer stating „I will work X hours a week for you“. It's not about the output.

If you work less than X hours, you commit fraud and breach the contract. Your employer can then fire you for an „important cause” and typical legal protections don't apply. If your employer forces you to work more than X hours without compensating you appropriately, he breaches the contract and you can sue him for that.

So obviously, there have been a lot of legal cases like „Does changing your clothes count as working hours“, „Does driving to a client count as working hours“, „How do working hours need to be tracked“,...

> healthcare loses their incomes as people become healthier not being obese -> jobs are lost there too, plus those same people will now have a lot of energy and losing their incomes.

In most developed countries, an effect like this would be desperately needed in order to prevent the health care systems from crumbling under the weight of few healthy, young people financing many old and sick people.

Worst case would be Pharma companies making a bit less profit, stable or falling health insurance costs and health care professionals having enough time for their patients. Not exactly a disaster.

Not having the bad stuff at home is the most effective fix for me.

I am capable of exercising enough discipline while I'm shopping at the grocery store.

But I will eat sweets if there's anything at home.

I can confirm that I can focus much better throughout the day, have higher mental clarity and energy if I do some sports in the morning.

It doesn't have to be the gym or a 45+ minute high intensity run - just jogging for 20 minutes and doing a few pushups or pullups is sufficient as „baseline activity“ for me.

Another good approach: Write down that email/ message in a text editor but just leave it there for a day.

Often, I'd realize that I don't want to send it at all. The emotion was processed and now it's fine.

Sometimes, there is a valid point (e.g. because someone was rude in tone) and I'd just give that as feedback in a non-violent manner.

> This moment will never come if you just sit and wait for it.

If you're literally just doing nothing, I'd agree.

But I haven't felt the urge to write anything creatively in over 8 years. Last year, I met someone (my muse?) who sparked this desire in me.

Nowadays I spend much of me free time writing songs, lyrics and I'm working on a draft for a novel.

It's not like I do that because I expect to make money from it - it's because I have to in order to feel al peace.

> I read these comments not as a defense of cheating, but as resignation that "This is just how the world is."

I'm also sympathetic with the cheaters. Everyone knows that these leetcode interviews are pretty dumb. Most people probably pass by memorizing everything and forgetting the stuff right afterwards. It's testing willingness to grind rather than ad hoc problem solving capabilities. If someone is capable of doing the job and just figuring out a way to get through the gatekeeping, I don't care. I know many people that cheated on exams at some point but are pretty legit otherwise.

> Look at this "personality type" as you call it: People who bullshit their way through life, self-promote, swindle, grift, fake it til they make it, hustle, exploit people's good intentions, take credit for the work of others, and so on.

There are a lot of psychological and societal mechanism that tend to punish this behavior. I'd rather be financially well enough off but honest rather than a super rich asshole.

> I'm one of those naive suckers who thinks hard work in the trenches should be rewarded. At work I often sit there doing the dirty gritty work that needs to get done, while peers write bullshit presentations and get face-time with the execs, and schmooze their way to promotion after promotion, and it sucks.

For myself, I decided that the best way is to be competent and do the hard work, but also to set boundaries, to ensure that my work is visible, to communicate ideas and feedback with execs and to fight against the fakers. You won't get anything if you don't ask for it. Taking down obvious fakes can take a while if the execs are too gullible, but it's worth doing.

If your execs are idiot assholes rather than competent people with some integrity, you should probably look for a job elsewhere.

> "The more unclear and early the product is, the more in-person time the team needs to grind together"

I agree and that statement from Sam makes sense to me.

Early on, it's great to work face to face. But once a project's or a Startup's direction is set, important decisions have been made and all leaders are on board, it's great to work remotely.

I'd say it's also good for new hires, especially juniors, to be on-site for a few months before going full remote.

Agreed and even using the same code for different levels can work well.

The juniors spot obvious errors. The seniors also spot logical flaws and the conversation quickly moves into "how could this be done better, overall".

I've had good experiences as an interviewer doing code review style interviews.

The candidate gets some code and a ticket description of what the code should do.

If he spots an issue, he can fix it right in the code or simply explain it.

I'd that testing the ability to read and understand code is both less stressful for the interviewee (no "judging every key stroke" stress) and more helpful for the interviewer.

Well, of course he's still more privileged than others but he definitely makes a point.

While rich kids do get an edge with private tutoring and other support from their parents, well designed standardized tests allow intelligent + hard working candidates without many resources to get in.

If only CV and grades matter, the system is way more rigged in favor of the rich kids.

Nah that's pretty unlikely. No leveraged trades/options/derivatives without some kind of accredition + a max order limit based on proof of funds/ income could happen, though.

> As a society, we should look out for the vulnerable part of our population.

Having read a bit about the neuropsychology of gambling addiction, I agree.

Slot machines are specifically designed to turn those who are susceptible to it into addicts.

Some people simply don't have the mental capabilities to distinguish a legit investment from a scam.

Maybe some kind of authorization should be required if you try to wire your life savings to a crypto exchange. But banks only ever ask questions if a lot of money comes in, not when it flows out :D

> They're exploitable because they're economically desperate.

2.5k € after taxes doesn't really count as economically desperate in Germany, I'd say.

Sure, housing is expensive - especially in Munich - but education and health isn't.

This is another hype-driven mania where some people gamble away their money. This guy could have just as well played roulette and bet 50k on a number.

I wonder if any kind of investor/debitor protection would help in these cases.

How did he get those loans? Why would a bank approve a 1.5k down payment with just 2.8k income after taxes, leaving just 1.3k for rent and food in Munich? Might be OK when buying a flat - but for gambling on the stock market?

Should more verification be required for risky or high volume trades? I just need to check some boxes with "yeah, I know that stuff" and I could trade derivatives + options with leverage.

> If you set up all the CI/CD pipelines to make that work.

I've been "abusing" GitLab CI pipelines a lot for running periodic background jobs that I'd like to have separated from the "regular" backend and without worrying much about deployment.

It supports Cron Syntax for scheduling, manual re-runs, provides secret management and download of artifacts, alerts me on failure and allows me to easily use the docker images in our registry.

Sure, I'm not storing anything in a database or calling the pipeline via an endpoint. But if that's needed, it should probably be "part of the regular backend".