arandr0x

I think it's worth mentioning that while biometrics for identification have flaws (as mentioned in this thread, they're not 100% collision free, are not necessarily secret, and are non-trivial to collect), DNA has a different risk profile if it leaks than fingerprints and iris scans if you consider technology advances. DNA could let people (moving beyond government, since you should probably assume anything in a government database will be leaked at this point) target your family and not just you, it includes information that could let adversaries find out ways you're uniquely exploitable (for instance allergies, sensitivities, diseases) and in general its potential for harm goes far beyond impersonation or being used in court.

(On the plus side, I suppose, I think the story on storing DNA at the scale we're talking about is not fully complete. DNA does denature and it takes a reasonably good sample to get a full genome sequence, and fully sequencing and storing data for every person has other practical issues. The article itself only references using DNA results to "prove or disprove biological sex", which is much more trivial and while it's likely to come with its own problems and edge cases, is also much less information.)

It's encouraging to see browsers actually deprecate APIs, when I think a lot of problems with the Web and Web security in particular is people start using new technologies too fast but don't stop using old ones fast enough.

That said, it's also pretty sad. I remember back in the 2000s writing purely XML websites with stylesheets for display, and XML+XSLT is more powerful, more rigorous, and arguably more performant now in the average case than JSON + React + vast amounts of random collated libraries which has become the Web "standard".

But I guess LLMs aren't great at generating XSLT, so it's unlikely to gain back that market in the near future. It was a good standard (though not without flaws), I hope the people who designed it are still proud of the influence it did have.

Is there not a confounding factor at play that a more functional government would facilitate both more democracy and more legislation on newer technology? Is this notion that "it might be nice to help your target market have a generally working government to facilitate them being willing to divert money towards non-corruption goals and able to protect your market with laws" really that new?

(Here the model would be that democracy is something that countries develop once they have some OK government systems, not that democracy in itself makes those systems better, but it works with the causation the other way too)

His parents might but I don't think they're accepting bribes from Stanford profs. Maybe CZ can advocate for him to have a rival again.

I'm finding it interesting (after spending time in the EU where ChatControl is front and center in the news) that there's been this resurgence of interest in electronic surveillance and information warfare suddenly in countries that 10 years ago spent a significant amount of the news cycle decrying the US's NSA and China's internet police. What's changed?

This is a Canadian bill and the party in power are (relatively) progressive and (definitely) democratic, though they are to the right of the American progressives and to the left of the American Democratic Party. More to the point, they do not remotely own everything.

They are unlikely to pass it directly without changes though. They are a minority government, and the Conservatives have been backed by foreign actors in various capacities in election cycles and are a strong enough opposition that they'll resent the crackdown, and the leftists are generally not fans of the local telecom industry (or the Liberal pandering to industry in general) and will probably want either less protection for telecoms or more protection for individuals.

This is extremely common throughout the world for businesses that sell alcohol or variations on gambling - and while I don't necessarily think the advertising should be _illegal_ (in those cases with non-controlled substances at least), I've always been shaken that the many people involved in it don't seem to see how it could be immoral.

I'm unfamiliar with the business of finding illegal immigrants. Isn't the idea of detaining them that you're worried they might flee if they know they're going to be deported? If so isn't it counterproductive for ICE to make noise (announcing themselves basically) if doing a 5am raid for farm workers? You'd think showing up at the field in a separate bus and spreading out would work best.

I think the issue you're having is that taking it way too seriously is exactly what creates the feeling of belonging. There's a number of forums or clubs where one might discuss morality, myth, creation stories etc more like wrestling - fantasy roleplaying groups and fantasy writing circles often do include those themes for instance. But it's harder to feel true belonging through discussing whether your bard is an asshole for sleeping with a gold-hoarding dragon if you're 100% aware that neither bards or dragons exist.

Is there anything you do take super seriously? The environment, your job, parenting, a specific hobby or something? Something you think other people are missing out on by not experiencing? Looks like a club based on that would work better for you.

Honest question: for women, are there any churches or other religious or spiritual traditions that do not limit them or afford them a second-class status relative to other members? (This can be in terms of church elders/leadership positions or just importance in activities, speeches giving women good role models, etc). For instance Catholic traditions usually provide quite a few female role models (not least Mary) but heavily sideline women in actual church operations, whereas Protestant traditions have more equitable leadership ranks, but don't give as much space to women who are not in leadership. I am not clueful enough to discuss other traditions sadly.

It would be interesting then for similar studies to add a dimension in addition to self-reported loneliness on self-reported time spent outside the home. While it's likely that sedentary elders are lonelier, it's not a 100% overlap (some elders live with close family and some lonely people are still very physically active/outside). You would expect lonely, active people to have lower death rates than lonely, sedentary people with similar pre-existing conditions under your hypothesis, and it would be a powerful thing to prove because it's a lot easier to make people go out than to make them make friends.

I found the take a little too much on the doomer side for someone who presumably has several years of management experience. Yeah there's been a lot of social media posts and talking about the efficiency era, AI slop etc but over there in the real world you're working with humans. Some are going to be operating under a shareholder- or investor-derived goal to improve margins and some are not, but even for those who are "improving margins" looks different at every business (depending on e.g. current headcount, COGS, whether you use contractors, etc). I feel like it's a super reductive take to go "aaah, this current culture is anti-human or anti-empathy" rather than like, look at the actual actions that are being taken, who is benefitting, and what specific negotiating room yourself and your team have in this value context.

I find the actual advice here very worthwhile, though.

It is possible to do leetcode without practicing, even before AI. That said the structure of the Big Tech process is also quite long/multi-step with many opportunities to give up during, which helps select for drive. It's hard to do this effectively with shorter processes. It is however always a good practice to 1) design very hard interviews but 2) give a lot of preparation to candidates beforehand, even for non-leetcode interviews, as it helps filter who can efficiently and diligently use provided information to increase their performance.

There is a reason religions tend to agree on 1) the worth of your life is not your own to judge and 2) what makes humans moral is they have free will - good or evil is found in the places you can control, even and maybe especially when it's hard.

Because otherwise you make morality a game and people are distracted from the deeds by the point system (The Good Place is another good TV show that has a funny satire of this).

Does EU-friendly mean timezones (GMT-4 to +3 I suppose), citizenship, or current residence? With this including contract roles it's a bit hard to infer. Asking because I really like the tech stack and there aren't a lot of point cloud roles around.

I think the post is getting at the idea that pedigree is not a reliable predictor of talent, but because it's a convenient and standard one, everyone uses it (which in turns reduces its usefulness). It's harder for a recruiter to fully experience the perils of hiring mediocre people, but they're definitely at ground zero for "what's on a resume is mostly not representative of actual talent".

That's definitely a valuable take, but it's worth noting that not everyone will make better decisions just by sitting through enough fires, and it's also possible that your good enough person will fail to notice some larger risk or market shift that the person you could've waited for would have, because they'd have seen it before.

Hiring decisions tend to be a hindsight is 20/20 proposition.

A more generalizable approach might be to consider - what are you looking for that most other companies either are actively putting off or passively neglecting, and what's the best way to identify the best engineers in that group. To use examples in the post, if you're remote then you can get "startup experience - hard worker - impressive project - aces your 20 ridiculous interviews" by getting in front of people who live in Ohio and people who live in the Bay Area and low key hate Caltrain. If you're willing to pay top of band salary all cash, ala Netflix, then you can be a Bay Area Only Senior Elites Need Apply type startup.

What about other things? What if you are, in fact, willing to let engineers decide whether they address tech debt, like the post calls out? Or, you don't overvalue confidence and talking and can appeal to female engineers, quiet engineers, or in general less competitive types? What if you want hard worker startup experience passes pseudo-IQ tests, but they don't need actual coding experience measured in years and you think AI and training can bridge the gap?

Note, I'm not saying any of these companies will necessarily be more successful with their hires, but they're being intentional with who they hire and how that fits the company's advantage in a way that the "you and everyone else" profiled in the post do not. Like, figure out what makes you different. Figure out how that will make your people different. Then write it in the job description, black text on white background (or the reverse in dark mode), plain language, so it's obvious.

I have thought about this because I've wanted such a platform as a candidate, and I think an interesting angle would be having the jobseeker's loved ones buy it for them as a gift. Many times, I've been sad that I wanted to "help someone " with their search but I did not myself have a role and didn't feel effective. Friends who are unemployed usually won't accept cash from you but they might accept a subscription to a job search service.

If you're good at 3d graphics or game programming I have a lead on a contract that could work (it's too much work to do by myself, client wants people in Canada, my network won't do it for less than 5 figures). If not your skill set, my point is reach out to other contractors, they may have overflow work or e.g. design/front-end contractors might need someone to refer for backend/devops work.

I would imagine shaming doesn't work because I think residential GCs have higher demand for workers than there is supply, but the cameras still solve the problems of making it easier for the GC to react when it happens (and the reaction could be offer to pay that sub more if the project is late or all the other subs have been showing up, realizing the work from the earlier stage wasn't done, and going home, or it could be lengthening their project schedule).

> we're heavily subsidized by huge grants and funding

this is how it is in Canada too! My city has a huge manufacturing sector so a lot of these little startups with super niche products that take lots of R&D are found there. But no one talks about us because engineers aren't paid doctor money here (the grants aren't THAT good, which I think in the US defense sector they are).

Software engineers have full-time managers (which are a lot more overhead to pay for but kinda serve that purpose from the client perspective) and are paid well enough and consistently enough to usually only work one job on a given day. Subcontractors sometimes do decide not to show up to your job site because another employer offered them a bonus to do theirs that day instead. The point isn't (only) to humiliate or do a show of power to the workers, it's to counter an economic incentive they have.

That said, for a lot of subcontractor trades, it's so hard to find anyone that I'd worry about the reverse: you get known as "the freaks with the cameras" and no one good bids on your stuff anymore, and then the delivery is even more delayed.

There are also "legitimate" reasons to fail first round interviews, such as the recruiter picking your resume for a role that you are not actually qualified for (which the hiring manager notices after the interview), your salary expectations being too high for that specific company size/geography/role (usually a communications mismatch where they advertise a more senior role than they actually have), or it coming out that you are not legally allowed to work for them (they cannot sponsor visas; they don't hire in your country; they require citizenship; etc).

All these cases are technically the company's fault, not the candidate's, but they do often come out right after the screening call.

I'm a manager whose definition of conflict mostly matches yours, so here's how I handle it (as an interviewee) and how to translate the "default behavioral interview question" to something you can answer:

- the scope of the question is usually "how do you reconcile your worldview, desires, and best interest with someone else's". When answering, I tend to first clarify to the interview that "conflict" is a word that has intense connotations for me, and then ask "Are you asking about a time I had to negotiate with a person that was coming from a different point of view or background, or about how I handled an emotionally charged situation?" (As a manager, it is more common that I do get asked about emotionally charged situations, which also take skills to defuse, just different ones.)

- At least for me, I often think of non-emotionally-charged disagreements as negotiations; there are absolutely negotiation skills to be used to resolve disagreements efficiently and effectively, and if you can't think of any, it may be something for you to practice. Do you often get what you want? When you do, is it mutually beneficial? How does that course of action happen?

- Sometimes you can sidestep the question and showcase better conflict resolution skills by thinking of a time you arbitraged a conflict between two other parties. Describe how you dealt with each party. Did you feel you were unbiased? If you were initially more on the side of one party, how did you set that aside to listen to the other person?

- Don't mention abuses of power (by you or the other party) in a conflict resolution question. Also don't mention conflicts that are primarily personal - the origin of the disagreement should be about work, not tone, physical threats, intimate history, etc.

- If you do have a story about building a working relationship with a known problematic person, so that your work with that person yielded good results and you didn't fall into the trap of a bad personal relationship with them, you can mention what you did to get there. E.g. "X was widely known for his temper and would often yell during code reviews. Although this caused problems with all his teams, and he doesn't work there anymore today, I always got civil, useful code reviews from X by sending him my points in writing in advance, and expressing that I valued his feedback which was why I wanted a trace of it."

In a reverse Turing moment, I actually submitted this joke to a chatbot, and it got it.

I've worked on the reliability team at a place that made what they call an "edgy app" (long before fly, so we made our own stuff on baremetal) and to be fair, reliability is the actual challenge when you're talking multi-region seamless deployments of scalable web platforms. It makes or breaks that kind of product at least for the original market.

Money doesn't contribute all that much to solving it though. It's usually time to make the early mistakes, continuous improvements, etc. But AWS had several decades to solve this, whereas fly clearly hasn't, so I'm not sure I'd count them out yet.

I'm not seeing how your example includes an "obvious improvement" that "only the best engineers would do". I would use arrays over lists but I have a ton of low-level experience so I tend to care about performance, but I wouldn't necessarily expect the "best" engineers to do so unless the prompt says "Please take special care to turn in the fastest solution".

There's a couple obvious tests (written unit tests) that I would expect the candidates to think of for their roman number sorting function, and I usually graded the take homes better when there was 1) evidence of testing 2) evidence of smart, appropriate edge case testing, but not having time to write polished tests in 1h isn't necessarily a red flag.

I'm really struggling to find how simple examples like this can demonstrate anything like GP mentioned.

Woah, I've been a hiring manager long enough that it's been a while since I've done a take-home code exam myself but I don't even think I would grade a LLM that way (because I moved to AI now). Unless your code exam is super trivial or the boxes themselves are table stakes ("code runs without errors", "code includes more than one function"), coding is creative enough that it's hard to come up with 20 checkboxes that cover whether a sample is any "good" let alone "shows better decision making".

I've had a few bad experiences when sharing feedback with candidates myself and I would understand doing the checkbox approach for feedback and/or just never sending detailed feedback, but actually grading submissions pass/fail based on a subset of criteria you jealously guard from candidates essentially selects for lucky people. If I wanted to do that, I'd just shuffle the submissions by number of bytes and discard everything that's a multiple of 5 or something.

I'm coming from a different place than most other HN commenters - I now work for a tech startup, but have a graduate degree in biochem where I worked in drug discovery. As a result, and before I decided I was not ethically congruent with working for Big Pharma, I had informational interviews, advisors, peers, and all kinds of honest interactions with people who work for pharmas, biotechs, CROs, research labs of all kinds, and hospitals. Of course, I wasn't trying to sell them anything.

You don't mention what your product or service is. If you're creating solutions on the per-project side, essentially starting a CRO, you should hire scientists with PhDs and talk to other CROs to see if they have more contracts than capacity (yes this is a lot like being a software contractor). If you're doing a product, "life sciences and pharma" is not a vertical, it's about a half-dozen of them. Pharma for instance has different problems in R&D and go-to-market, and it's typically not the same organizations that do pharma R&D and market activities. Smaller pharmas, CROs, academic research labs, and specific innovation departments might do R&D, and larger pharmas have armies of representatives selling different drugs (mostly) to doctors or hospitals, marketing arms that organize and fund studies and clinical trials, M&A specialists that choose compounds to buy and market, subcontractors that produce the drugs in marketable quantities (this is not the same as e.g. the quantity you need for a clinical trial), and regulatory affairs staff, and all those people largely are ignorant of the problems that are not at their specific level of the value chain.

Once you have chosen a specific segment to focus on (e.g. you want to reach an early stage pharma that does have staff and funding and maybe is starting one clinical trial), you can reach them through several channels. One is asking your investors, your company's CEO, your incubator or other startup-adjacent staff for intros to people who are doing or funding companies like your target profile (be specific). You can also go to networking events at incubators, chambers of commerce, or startup-focused conferences for that stage. You will meet executives and investors, so you can ask executive-and-investor generic questions to have insight on market dynamics, funding situations, and growth stage. ("Where do you want the company to be in 3 years?", "What was your last executive hire, and why did you pick their profile?", "Who are your competitors?") This will not help with product but might help with pricing, time-to-market, and other elements of the "sales" side of the product strategy. At small companies where the exec is likely to approve every line item, you can also ask if they are customers of X competitor of yours, which will give you an idea of that competitor's market.

Another option is to look at who they're hiring. Staff at companies in this industry tend to have graduate degrees and their entry level jobs are as laboratory or project management staff, so you can try to find students at the tail end of their PhDs or postdoc at research centers with close industry ties. Calls with them should be easy to get and you can ask them about the technical jobs to be done your product is supposed to help with. Depending on what your product is, they may also have used something like it and talk about competitors, if your product serves a function that also exists in academia. Some professors can also be very good networkers (like investors) and if you can get a warm intro to them, they may spontaneously refer you to people.

Look at competitors' advertising. In this space it will be in trade journals, trade shows or conferences, and Internet-based advertising like AdWords and such. Find the ads themselves, follow the companies on social media to be sure to catch any public posts they make, and do your own ads in the same space to try to get 1) a profile on the kind of people who click them and 2) maybe email addresses of very curious parties.

At conferences (healthcare or otherwise) you're unlikely to have people committed enough to give you contact info for a follow-up call, but targeted conferences are great places to rapidly have 100+ 5-min shallow conversations. Walk on the floor in the dedicated vendor space, ask everyone you see who they work for and if they've heard of $market_leader in your tool space or $problem_you_solve. Half of them will not, that's ok. Ask those who do for a quick sentiment on $problem, if it affects them personally, and why/why not. Only ask (or beg) for a follow-up call if they look very interested and there's a good match between their space and yours, and if you do ask for a call, mention you have nothing to sell and are after information. You need to find the right conference or trade show for this. Things your competitors advertise at (better: sponsor), or even ask your existing customers which they go to. Health conferences specifically are about 50% doctors, so that's why hanging out in the vendor area is usually a better bet. Other vendors will also have downtime and you can talk to them when they look bored.

Cold emails/LinkedIn contacts/phone calls work too, but at like 1/100 the rate of contacts at conferences or warm intros. So, if you have a couple afternoons free and don't mind the occasional very vocal rejection, you can do that, same deal as at conferences with starting with a short intro&question to qualify who they are in the space and asking politely for an info call only if there's a relatively positive response.

I mentioned phone calls. The reason Airbyte's strategy worked for them is software developers live on their computers. Most people in other industries don't. Some are even (gasp) over 50 and don't have a ton of downtime at home due to family and other obligations. If you're doing warm outreach, and your contact gives you both an email and phone number, call. If you get a business card with both an email and phone number, call. The only time you shouldn't at least try calling is if you got the number by paying for a database or otherwise have very unqualified leads. Never hold people on the phone too long unless they're the ones talking: if there's a positive response to outreach, schedule a follow-up call.