- A 10 min visit to a specialist office, where I was seen by a nurse practitioner who was able to refer me to an exam without involving the main physician, costed $800 (just the visit, not the followup exam), which also went through deductible.
- A routine colonoscopy came at $19k with bills that kept coming for the next 6-12 months from various parties. Consumed my deductible in one shot as the first bill came.
- A visit to the ER that lasted 10 mins with an XRay costed me $5k, also sent in tranches for the next 6 months, to the point where I thought I was victim of some scam (I wasn’t). Maxed my deductible as well.
I have a couple non-critical health concerns right now I’d like to consult a doctor about, but I won’t drop another huge deductible just for that.
I am a dual citizen from a European country with a high quality healthcare public system (but admittedly longer waits than the US ). I’m not eligible for it since I’m not a resident anymore (I will when/if I move back), but as I went visiting family I booked a private appointment with a well known cardiologist and he spent a good 50 minutes with me, for a total of EUR 100.
$80-300 in capital of Poland when going full private. Its crazy than not everyone in US is combining medical/dental procedures with Euro/Asia vacation yet.
Also, the quality of service is top notch. Not just the procedure execution but in private dental clinics (in Gdańsk) you lay down on a bed and there's a sizeable OLED screen at the ceiling with slow-mo clips sporting waterfalls, beaches, mountains and other "nice views" you'd expect from a deep house music clip. The background music is elevator-like though.
Every time I start doing some network packet-based troubleshooting, for example firing up tcpdump to debug a tcp window size improperly tuned that caused a performance degradation over a high latency link (this is a random example I did last week), coworkers look at me like I pulled off some sort of wizardry. I feel it’s knowledge that most SWEs working on distributed systems should have acquired early and routinely use, but clearly that’s not the case.
I personally think it’s better to struggle early on than to give away money like candies to founders with dubious businesses who will then trick gullible employees into thinking that these companies’ massive valuations mean anything. I kid you not one founder told me once my equity was safer in the company than liquidated and put in the SP500, when I inquired about tender opportunities.
I have vested equity (that I painfully exercised) in a “2020-era unicorn” that has been a complete zombie ever since. I wish they’d just go out of business for me to at least declare a capital loss on my taxes.
I just have myself to blame clearly, so just stating an opinion.
>> I personally think it’s better to struggle early on than to give away money like candies to founders
I agree, and disagree.
My business was bootstraped, we never took outside investment, and it succeeded and makes money. It employs 50 odd people and is a big fish in a fairly small pond.
We make the world better for a few thousand people (customers).
But our approach couldn't lead to a Facebook, or Amazon or Uber etc. Products like that work best once they achieve scale, and achieving scale is expensive. But Uber (goes example) works poorly with 50 drivers in 1 city.
For most businesses and most founders, more is achieved with less. Most businesses don't need scale to be effective.
But equally, the VC approach is necessary for some subset of problems.
You want to sell that equity to create a loss? If it’s worth zero, I’m happy to go through the exercise if it’s marketable. This offer is only good if it’s actually worthless and I’m not stiffing you on the value.
I would simply hold it until the value goes to zero. In the event enterprise value doesn’t go to zero, I would gift any gains back to OP after taxes (because I’m trying to do them a favor, not make an investment).
Staying years at an early startup in exchange for paper money (and having to pay AMT + exercise costs) is one of the biggest regrets in my career, I truly don’t know what I was thinking.
I still have a bunch of this paper junk and I wish they’d at least go to $0 so that I could claim a capital loss on my exercised cost basis, but they are stuck in zombie land.
Genuine naive question: when it comes to Google HN has generally a negative view of it (pick any random story on Chrome, ads, search, web, working at faang, etc. and this should be obvious from the comments), yet when it comes to AI there is a somewhat notable “cheering effect” for Google to win the AI race that goes beyond a conventional appreciation of a healthy competitive landscape, which may appear as a bit of a double standard.
Why is this? Is it because OpenAI is seen as such a negative player in this ecosystem that Google “gets a pass on this one”?
And bonus question: what do people think will happen to OpenAI if Google wins the race? Do you think they’ll literally just go bust?
Most of us weren’t using Gemini pro models (1.0, 1.5, 2.0) but the recent 2.5 pro is such a huge step up. It’s better than 3.7 sonnet for coding. Better than o1, o3-mini models and now o3 and o4-mini. It’s become my daily driver. It does everything I need with almost 100% accuracy, is cheap, fast, 1 million context window, uses google web search for grounding, can fetch YouTube video transcripts, can fetch website content, works in google workspace: Gmail, Docs, Sheets.
Really hard to beat this combo. Oh and if you subscribe to their AI plan it comes with 2 TB drive storage.
Maybe because Google is largely responsible, paying for the research, of most of the results we are seeing now. I'm not a Google fan, in the web side, and in their idea of what software engineering is, but they deserve to win the AI race, because right now all the other players provided a lot less than what Google did as public research. Also, with Gemini 2.5 PRO, there was a big hype moment, because the model is of unseen ability.
Maybe they deserve it but it would be really bad for the world. Because they will enshittify the hell out of it once they're established. That's their MO.
I don't want Google to have a stranglehold over yet another type of online service. So I avoid them.
And things are going so fast now, whatever Google has today that might be better than the rest, in two months the rest will have it too. Of course Google will have something new again. But being 2 months behind isn't a huge deal. I don't have to have the 'winning' product. In fact most of my AI tasks go to an 8b llama 3.1 model. It's about on par with gpt 3.5 but that's fine.
The situation with LLMs is much different than search, Google doesn't have such a large lead here. LLMs are social things, they learn from each other, any provider with SOTA model will see its abilities leaked through synthetic training data. That's what GPT-4 did for a year, against the wishes of OpenAI, powering up millions of open model finetunes.
Gemini is just that good. From my usage it is much smarter than DeepSeek or Claude 3.7 Thinking models.
A lot of Google’s market share across its services comes from the monopoly effects Google has. The quality of Gemini 2.5 is noticeably smarter than its competitors so I see the applause for the quality of the LLM and not for Google.
I think it’s way too early to say anything about who is winning the race. There is still a long way to go; o3 scores highest in Humanity’s Last Exam (https://agi.safe.ai/) at 20%, 2.5 scores 18%.
As a googler working in LLM space, this feels like revisionist history to me haha! I remember a completely different environment only a few months ago when Anthropic was the darling child, and before that it was OpenAI (and for like 4 weeks somewhere in there, it was Deepseek). For literally years at this point, every time Bard or Gemini would make a major release, it would be largely ignored or put down in favor of the next "big thing" OpenAI was doing or Claude saturating coding benchmarks, never mind that Google was often just behind with the exact same tech ready to go, in some cases only missing their demo release by literally 1 day (remember live voice?). And every time this happened, folks would be posting things to the effect of "LOL I can't believe Google is losing the AI race - didn't they invent this?", "this is like Microsoft dropping the ball on mobile", "Google is getting their lunch eaten by scrappy upstarts," etc. I can't lie, it stings a bit when that's what you work on all day.
2.5 was quite good. Not stupidly good like the jump from GPT 2 to 3 or 3.5 to 4, but really good. It was a big jump in ELO and benchmarks. People like it, and I think it's just psychologically satisfying that the player everybody would have expected to win the AI race is currently in the lead. Gemini finally gets a day in the sun.
I'm sure this will change with whenever somebody comes up with the next big idea though. It probably won't take much to beat Gemini in the long run. There is literally zero moat.
I dislike Google rather strongly due to their ad-based business model, and I was previously very skeptical of their AI offerings because of very lackluster performance compared to OpenAI and Claude. But I can't help but be impressed with Gemini Pro 2.5 for "deep research" and agentic coding. I have subscriptions with all three so that I can keep up with SOTA, but if I had to choose only one to keep, right now it'd be Gemini.
That said I still don't "cheer" for them and I would really rather someone else win the race. But that is orthogonal to recognition of observed objective superiority.
I think a lot of us see Google as both an evil advertiser and as an innovator. Google winning AI is sort of nostalgic for those of us who once cheered the "Do No Evil"(now mostly "Do Know Evil") company.
I also like how Google is making quiet progress while other companies take their latest incremental improvement and promote it as hard as they can.
I think for a while some people felt the Google AI models are worse but now its getting much better. On the other hand Google has their own hardware so they can drive down the costs of using the models so it keeps pressure on Open AI do remain cost competitive. Then you have Anthropic which has very good models but is very expensive. But I've heard they are working with Amazon to build a data center with Amazons custom AI chips so maybe they can bring down their costs. In the end all these companies will need a good model and lower cost hardware to succeed.
2.5 Pro is free, and I'm sure there's a lot of people who have just never tried the best models because they don't want to pay for them. So 2.5 Pro probably blows their socks off.
Whereas, if you've been paying for access to the best models from OpenAI and Anthropic all along, 2.5 Pro doesn't feel like such a drastic step-change. But going from free models to 2.5 Pro is a crazy difference. I also think this is why DeepSeek got so much attention so quickly - because it was free.
The key is Gemini being free through AI Studio. This makes their technical improvement more impressive when OpenAI sells their best models at ridiculous prices.
If Google engages in price dumping as a monopolist remains to be seen but it feels like it.
The LLM race is fast paced and no moat has developed. People are switching on a whim if better models (by some margin) show up. When will OpenAI, Anthropic or DeepSeek counter 2.5 Pro? And will it be before Google releases the next Pro?
OpenAI commands a large chunk of the consumer market and they have considerable funds after their last round. They won't fold this or next year.
If Google wants to win this they must come up with a product strategy integrating their search business without seriously damaging their existing search business to much. This is hard.
Because now it has brought real competitions to the field. GPT was the king and Claude had been the only meaningful challenger for a while but OpenAI didn't care about Anthropic but just be obsessed with Google. Gemini took a quite time to set the pipeline so initial version was not enough to push the frontier; you remember the days when Google released a new model, OpenAI just responded with some old models in their silo within a day only to crush them. That does not happen anymore and they're forced to develop a better model.
A lot of the negativity toward Google stems from the fact that they're the big, dominant player in search, ads, browsers, etc., rather than anything that they've done or any particular attribute of the company.
In AI, they're still seen as being behind OpenAI and others, so we don't see the same level of negativity.
I prefer OpenAI and Anthropic big time because they are fresh players with less dominance over other aspects of digital life. Not having to login to an insidious tracker like Google is worth significantly worse performance. Although I have little FOMO here avoiding Gemini because evaluating these models on real world use cases remains quite subjective imo.
I like to think I am extremely attentive to bug reports (I’m always paranoid about my code possibly containing bugs, so I err on the side of turning every stone), so I would probably not make Jerry’s mistake here, but sure as hell if a manager tried to humiliate me that much I would just quit on the spot.
In my opinion, there are ways to share feedback that allow another person to save face, letting them process it on their own terms instead of pounding them like this in a single session until they are “defeated”.
Such feedback can then be politely repeated, if the issue reoccurs later on, and formally documented as part of a performance warning, simply letting the other person know, once again without insisting, that this is a serious behavioral issue that will have repercussions if not actioned, and that you are there to provide any context should they want to talk about it more.
That is, in my opinion, a way for a leader to show that every team member is treated as an adult and responsible for their own actions and outcome.
To be honest, reading the communications, the manager backed him in a corner and then just kept hammering. There's no place for the employee to go besides to defensive.
If you're a manager and the employee can't give you anything but complete and utter capitulation to satisfy you, you're not taking to them, you're just making yourself feel better about disliking them
> letting them process it on their own terms instead of pounding them like this in a single session until they are “defeated”.
When you combine this with a difference in authority, as it is in this situation, it's even worse to behave this way. If this person was a peer and wasn't forced to sit through this "meeting" I can imagine they would have hung up about 20% through it.
He treated his subordinate like a child and didn't even handle it well in that context. Even in the situation where the employee was is wrong it reflects poorly on the manager. It's wild to be bragging about it on the internet. Being the authority figure means you are expected to be the adult in the room and understand the other person even when they lack the words to express themselves. That's why you're paid the big bucks.
The manager was hounding Jerry because (at least according to their version of the story) he was still not accepting responsibility for his actions and acknowledging that what he did was wrong.
If Jerry was not willing to accept that responsibility, then him quitting would be the ideal outcome anyway, so I understand why the manager was pushing him like this.
OTOH the manager should have been smarter about acknowledging Jerry's thinking (technically Bermuda is not in the Caribbean) and in explaining why they needed to overrule that (vacationers searching for Caribbean destinations mostly want to see listings for Bermuda). Their messaging was pretty disempowering to Jerry, kind of like saying "shut up code monkey, dance when the product team tells you to".
> every team member is treated as an adult
Adults take responsibility for their own actions. If someone is lying like this and being difficult to work with you want them to either get it together or leave. It's not about saving their feelings or letting them save face anymore. Let them leave. You can replace Jerry, and you'd rather he leave than the people who aren't being difficult.
Eh, even when you don't like it, managing egos is an important part of being an effective leader. We're social creatures, and nobody wants to work with somebody that is comfortable humiliating them.
Yeah, even just acknowledging the other guy's valid point, like "I agree with you, Bermuda is not part of the Caribbean, so yes it's deceptive to return it in the search results."
Sales is often deceptive. I think that's why the author was so defensive, notably about "honesty".
>>> Jerry: But I made some good decisions about what shows up when. For instance, I don’t like the idea of showing deals for Bermuda when someone searches for “Caribbean”
While Bermuda not being part of the Caribbean might be true…it’s really not a valid point in this circumstance. It was an apparent dev assumption that this level of precision was desired even if it wasn’t requested or its absence as a requirement was a mistake. When it was reported as a bug, that should have prompted the developer to clarify with the PM the OG requirement so they were both on the same page. He didn’t, he assumed she was wrong…apparently twice.
The PM probably understood that to their lay users the terms “Bermuda”, “Caribbean”, and “Islands” all have a degree of marketing relevance to each other (There is a reason why you might dumb down search results for something like that). Judging by what was information given in the story, apparently dumbed down search results was known and required in the past.
Not sure there is any “right” way to gently coach an employee that disregarded proper protocol and requirements, lied about it, blamed a colleague, and initially refused to accept responsibility when it was brought to their attention. That would be damn frustrating. I know HN loves to blame the manager, but this is one of those instances where it seems every portion of the problem and its escalation appears to lie solely at the feet of the developer. It least in my opinion.
It’s super easy to coach an employee about this. I’ve done it many times, even once publicly to a director while I was a senior. Just acknowledge their train of thought extra generously, then explain why this case is different. ChatGPT has been trained to do this. I’d say your paragraph 2 was great, and yes the manager should have tried talking about “level of precision” or “marketing relevance” instead of defaulting forcefully to “because i/product said so”.
Yeah I agree that the manager seemed to leap pretty quick onto the “you don’t get to have independent thought in secret” train. Of course we only see one side of the story too. But it kind of feels to me like this might not have been the first time there has been a need to coach this person either. Not sure that just one instance really gets you to lack of trust.
I’ve been in the business a long time in multiple roles up to senior leadership. I am at the sunset of my career and found a lovely tech role where I happily don’t have to manage anyone but myself and it’s glorious. In all that time I have encountered my fair share of the Jerry types in the last almost 40 years so I am probably jaded and know the frustration.
I’d suggest that if the general populace that use your app sort of see “Caribbean”, “Islands”, “Bermuda” to a degree of equivalence to the general area that they are considering for vacation, it’s probably not a dark pattern.
Ethics is like wearing a condom. If someone tries to tell you that you don't need to worry about it (after defaulting to force!), then you *really* need to worry about it.
Edit: Sure enough, look at your comments. Elon, Tesla, Musk, Trump, Trump. Hilarious + predictable.
> I know HN loves to blame the manager, but this is one of those instances where it seems every portion of the problem and its escalation appears to lie solely at the feet of the developer.
Yeah, that is a tell that the manager is being unreasonable. What are the odds that every portion of this sits with the developer? Close to 0. That isn't how miscommunications generally happen. But it is a pattern that turns up when borderline abusive people report their interactions on the internet [0]. I don't trust this paraphrase of what Jerry was saying at all. It seems much more likely he led with something reasonable "Bermuda isn't in the Caribbean and we made a decision in the past to exclude Caribbean deals when that sort of search happens" then the manager fumbled the conversation in follow-up and managed to make it weird.
Though it does seem like the developer was making mistakes; I've seen scenes play out that I'd rate as similar. Part of the dev's job is to manage their manager when that manager is struggling. In this case the manager seems to have gotten stuck in the mindset of being a big monkey and the response to that is to let them monkey out peacefully and not make a fuss.
It could have been worse; but this looks like bad-to-average leadership on display.
[0] Big red flags in the "I wasn't sure he was working in good faith" and "Maybe he was feeling lazy" quote with no evidence. We're likely seeing misreported conversation; the manager doesn't understand what is going on conversationally if he got to that mindset in this sort of situation. What else did he get wrong?
While it’s certainly written from the managers POV and will show that bias, but the behavior and attitude he is describing is not unheard of from SWEs and the personality types that are attracted to that vocation. Frankly it’s frustrating and a challenge that cannot be easily dismissed as “bad-to-average leadership”. It might be bad leadership, but it’s bad behavior from the employee too and that should be acknowledged.
When I managed technical people I used to describe the challenge to management folks outside of tech that in tech management you are managing a group of people where everyone is absolutely convinced that they are smarter than everyone else around them. Often not only just smarter but their other colleagues are bumbling idiots.
So my bias from experience is that this dev was wrong, because I have been through very similar scenarios with similar Jerry’s. I spent 2 decades managing engineers, but haven’t for a decade so I acknowledge that times change and perhaps stroking fragile egos is a management necessity in 2025 where it wasn’t in 2015. I also had the benefit of supportive executive leadership that allowed me to quickly cycle out those personalities like “Jerry” to build a team where even if folks were convinced they were smarter than their peers, they would at least communicate with each and be respectful of the entire team’s contribution that allowed us to minimize source issues like was described in this story.
I might expect many managers to be in alignment with Mr. respectfulleadership.substack.com on similar logic to the principle that about half of managers are below the median manager in communication skills. Empathy is a rare and difficult ability. But if this manager was adept at it he would be doing a much better job of articulating "Jerry"'s actual concerns and reflecting on why he was failing to communicate with him instead of setting up and gloriously defeating the strawman we see in the article. While failing at communication so hard he felt he had to do the ticket himself; we might note.
There are engineers convinced that they are smarter than everyone else but that isn't what is happening in this vignette at all. Even with my grave concerns about how accurate the reporting is of what he said, half the story is the reportee trying to figure out what he's done wrong and how to stop his manager from beating him with a rhetorical stick. That isn't the action of a man convinced he is the smartest guy in the room. But it is the reporting of a manger who believes beating someone else with a rhetorical stick makes him look good.
I'm not even saying this guy is a bad leader overall; at least he isn't being especially passive aggressive and this doesn't say anything about him at his day-to-day. But if he understood what attitudes and behaviours he was displaying in this story he wouldn't have been so keen to publish it. People who've had to deal with abusive people in power positions are going to have alarm bells going off reading this. It looks like someone lying to themselves and the reader in order to feel good about wielding power. Although, reading charitably, it might just be ignorance and low-grade communication ability.
> There are engineers convinced that they are smarter than everyone else but that isn't what is happening in this vignette at all.
It’s the underlying cause though. According to what was written, Jerry made an assumption without any real investigation, didn't clarify the problem by engaging in communication to Sonya, argued his “superior” knowledge because of his assumption of her mistake, then apparently was dismissive again after Sonya apparently developed enough detail to escalate the problem to the leader who had to investigate and fix the issue himself.
Remove the Jerry’s arrogance all the way to the point of that rhetorical stick is there even a blog post here? Perhaps frustrated leadership is the effect, but Jerry’s poor attitude, handling, and arrogance is damn sure the cause.
There is scant evidence that it is the underlying cause. The manager didn't do his due diligence in figuring out what the cause was; I repeat myself but it looks suspiciously like he's disregarded whatever Jerry actually said, substituted a strawman and dealt with that strawman harshly.
At no point did Jerry assert superior knowledge. He should probably have talked to Sonia... but that isn't something that the manager seems to have picked up on. That is something that manager should have done, in fact, which is prompt the developer and Sonia to talk to each other. That seems like a much more reasonable underlying cause of problems than arrogance; a relatively minor mistake in failing to check what the ticket raiser wanted on a call when the ticket didn't seem plausible. If the manager had suggested that at any point he'd get a free mark for having done some good managing.
> Remove the Jerry’s arrogance all the way to the point of that rhetorical stick is there even a blog post here?
My guess is that is why it ended up being flagged; there isn't really a blog post here that is good to read. The post is basically Jerry mishandles a ticket in a minor way and his manager responds by melting down, mishandling the situation and making Jerry eat the heat for that in a style that looks mildly abusive. But the manager doesn't know what respectful communication actually looks like so he posted it expecting a warm response. Sure, Jerry made mistakes here but it isn't the elephant in the room.
If the manager had good communication skills he could have resolved this by articulating to Jerry some basic & polite actionable feedback instead of the display we actually got in the blog. There wasn't enough legwork done to justify a hard conversation.
I used to live nearby and my favorite “urban” hike was going up Glen Canyon, up and down the two Twin Peaks, loop through Sutro Forest and then go back to downtown walking down 17th. Loved those weekend walks.
Ha yes love that loop, so strange to see someone else tying it all together! Not quite as good but when I lived in the sunset — grand view to golden heights along the ridgeline then down to Laguna Honda trails and up to twin peaks and out sutro forest to Parnassus.
I read the post and I am thankful that these people exist, most of my life is made possible through the selfless work of wonderful open source developers like this person.
Sometimes I wish I was so passionate, whereas my philosophy in life towards strangers is a much simpler “fuck you, or pay me”. It allows me to sleep fairly well at night.
I have a few toy projects on GitHub, a couple of which gained a tiny bit of popularity, and I simply ignored every new feature request that I didn’t need, and especially those large PRs “here, I refactored your code to make it functional and it’s so much better now”. I simply said, with no regret: “I won’t merge this, feel free to fork the project, if it’s better I might even switch myself to your project!”. Some got mad, but I truly and genuinely couldn’t care less.
I’ve been through so many shenanigans during my previous life as a naive startup employee: paid huge amounts of AMT (which took years to recoup via AMT credits), was not offered 83b election, had to write huge checks to exercise ISO, had to pay taxes when exercising NSO, etc., but I had never heard of a company threatening to forfeit the RSU if tax is not wired to them, it’s simply wild, especially when the liquidity is so close.
If I was in your shoes and the amount was substantial, I’d consult a lawyer. I hired one to help me facilitate a secondary sale transaction contract and it cost me $4k, money well spent. Tons of them in the Bay Area.
I would truly love to know how this will end up for you.
I am sorry that you had to go through such shenanigans. Consulting a lawyer is also what we are working on right now. Asking just in case: do you have any lawyer to recommend? Thanks!
In case you don’t have much experience working with lawyers, start your conversation with them by saying, “The outcome I want is _________________.”
If you don’t do this, the lawyer will likely talk about many options, none of which will match your desired outcome. They also generally charge by the hour so you’ll be paying to hear about things you don’t care about.
The 83b election mechanism is a mess, and plenty of startups don’t explain it well to their employees, but I wasn’t aware that an employer had any particular say in it. Generally, the employee makes the election by mailing the appropriate documents to the IRS.
that AMT struggle is real, painful, and needs legislative changes when it comes to ISO exercise in private companies. especially when you have to set up a payment plan for the taxes owed, which are ever increasing due to fines and penalties, while the owed credit is ever losing value due to inflation and the lost opportunity to put that money to work. and on the state tax side you should plan on not moving and losing a job in that state as if you dont have any taxable income there you are shit out of luck on getting your money back. my only advice is that if you are offered a large chunk of options without an 83b and have faith in the company, exercise as soon as possible before the value goes up.
The only legislative changes we've gotten recently were actively designed to screw startups (thank the Republicans for Section 174 and a bunch of layoffs), so don't hold your breath.
- A 10 min visit to a specialist office, where I was seen by a nurse practitioner who was able to refer me to an exam without involving the main physician, costed $800 (just the visit, not the followup exam), which also went through deductible.
- A routine colonoscopy came at $19k with bills that kept coming for the next 6-12 months from various parties. Consumed my deductible in one shot as the first bill came.
- A visit to the ER that lasted 10 mins with an XRay costed me $5k, also sent in tranches for the next 6 months, to the point where I thought I was victim of some scam (I wasn’t). Maxed my deductible as well.
I have a couple non-critical health concerns right now I’d like to consult a doctor about, but I won’t drop another huge deductible just for that.
I am a dual citizen from a European country with a high quality healthcare public system (but admittedly longer waits than the US ). I’m not eligible for it since I’m not a resident anymore (I will when/if I move back), but as I went visiting family I booked a private appointment with a well known cardiologist and he spent a good 50 minutes with me, for a total of EUR 100.
reply