2,253 karma · joined July 8, 2015
Claude provides nicer explanations, but when it comes to CoT tokens or just prompting the LLM to explain -- I'm very skeptical of the truthfulness of it.
Not because the LLM lies, but because humans do that also -- when asked how the figured something, they'll provide a reasonable sounding chain of thought, but it's not how they figured it out.
Yes, gemini loops but I've found almost always it's just a matter of interrupting and telling it to continue.
Claude is very good until it tries something 2-3 times, can't figure it out and then tries to trick you by changing your tests instead of your code (if you explicitly tell it not to, maybe it will decide to ask) OR introduce hyper-fine-tuned IFs to fit your tests, EVEN if you tell it NOT to.
You're lucky to afford the luxury not to atrophy.
It's been almost 4 years since my last software job interview and I know the drills about preparing for one.
Long before LLMs my skills naturally atrophy in my day job.
I remember the good old days of J2ME of writing everything from scratch. Or writing some graph editor for universiry, or some speculative, huffman coding algorithm.
That kept me sharp.
But today I feel like I'm living in that netflix series about people being in Hell and the Devil tricking them they're in Heaven and tormenting them: how on planet Earth do I keep sharp with java, streams, virtual threads, rxjava, tuning the jvm, react, kafka, kafka streams, aws, k8s, helm, jenkins pipelines, CI-CD, ECR, istio issues, in-house service discovery, hierarchical multi-regions, metrics and monitoring, autoscaling, spot instances and multi-arch images, multi-az, reliable and scalable yet as cheap as possible, yet as cloud native as possible, hazelcast and distributed systems, low level postgresql performance tuning, apache iceberg, trino, various in-house frameworks and idioms over all of this? Oh, and let's not forget the business domain, coding standards, code reviews, mentorships and organazing technical events. Also, it's 2026 so nobody hires QA or scrum masters anymore so take on those hats as well.
So LLMs it is, the new reality.
There's always been oscilations, true, but the rate o change and trend on those oscilations is the real issue.
I think the models have reached that human training data limitation a few generations ago, yet they stil clearly improve by various other techniques.
My experience is claude (but probably other models as well) indeed resort to all sorts of hacks once the conversation has gone for too long.
Not sure if it's an emergent behavior or something done in later stages of training to prevent it from wasting too many tokens when things are clearly not going well.
I want to highlight this realization! Just because a model says something cool, it doesn't mean it's an emergent behavior/realization, but more likely post-training.
My recent experience with claude code cli was exactly this.
It was so hyped here and elsewhere I gave it a try and I'd say it's almost arrogant/petulant.
When I pointed out bugs in long sessions it tried to gaslight me that everything was alright, faked tests to prove his point.
Also, we can't survive an asteroid crash/extinction event with solar.
Nuclear is transcedental. If we had practically unlimited fusion power, we could build underground, grow plants in aquaponics and aeroponics and ride it out in underground cities and farms.
I've been thinking about this a lot lately in another context -- vira priests being anti-vax and realized it's the other way around: their motivation doesn't matter, but the viewers don't want to see moderate content, they want to see highly polarized and controversial topics.
The same with the claims about AI. Nobody wants to hear AI boosts productivity in nuanced way, people either want to hear about 10X or -10X so the market dictates the content/meme.
My issue is I've had my work laptop wiped twice because of things I've installed on it and it's a hassle to switch accounts/devices but I've love to give sprites a go.
I'm envious of those true believer kind of people.
My father is one of them and he's held absurd ideas as 100% facts and we've had many nasty quarrels about it, BUT it also means he 100% believes in whatever his current goal is and he's achieved a lot more than I ever will because he's unwavering in his beliefs and goals, whereas I'm always doubting and second guessing.
I also think that's the case, but I'm open to the idea that there are people that are really really good at this and maybe they are indeed 10x.
My experience is that for SOME tasks LLMs help a lot, but overall nowhere near 10x.
Consistently it's probably.... ~1X.
The difference is I procrastinate a lot and LLMs actually help me not procrastinate BECAUSE of that dopamine kick and I'm confident I will figure it out with an LLM.
I'm sure there are many people who got to a conclusion on their to-do projects with the help of LLMs and without them, because of procrastination or whatever, they would not have had a chance to.
It doesn't mean they're now rich, because most projects won't make you rich or make you any money regardless if you finish them or not
I don't like to be mean, but I few weeks ago the guy bragged about Claude helping him do +50k loc and -48k loc(netting a 2k loc), I thought he was joking because I know plenty of programmers who do exactly that without AI, they just commit 10 huge json test files or re-format code.
I almost never open a PR without a thorough cleanup whereas some people seem to love opening huge PRs.
Two caveats: I'm talking about early 2000's Romania where it was quite rare to see really muscular men AND I was in my early 20s, I have no clue if men in their 30s or 40s get the same kind of attention, it was the shallow kind of attention.
By the time my 30s hit I was already married, however we did have a very rough patch where we were very close to divorce, I traveled alone, was part of an NGO and I think a lot of the young women there got a strong clue about how rough things were between me and my wife because I also got a lot of attention from women back then but then it wasn't because of physique - it was because I was their photographer and high status in the NGO and much higher net worth.
Going the traveling groups, I also did get quite a bit of attention, because there were a lot of single women in their late 20's and I'm sure a lot of them got this fantasy they would find their soul mate on such a trip and I was 100% not interested in anything romantic but I was very open to socialize. All sorts of un-intended "adversarial" techniques I noticed worked -- again my portrait photography hobby seemed attractive to the ladies, when I noticed one was too interested, I pulled away so she won't get the wrong idea, which made her even more interested, I also chatted up foreign women in the hostels because I was geniuinely curious about their background -- I had a long conversation with a hot, early 30s phd in fish farming and a young woman in our group started interrupting us by saying "You know, we've been together in this group for 3 days and you haven't spoken to me as much as you did to this foreigner, why?". I clearly remember I talked to a psychology major woman in the bus because I'm very interested in psychology and I did quite a few years of psychotherapy and I'm very interested in Schema Therapy. At some point we discovered we run in the same park and she invited me to run together. It was then that it hit me she probably had the wrong idea, but I felt it would be weird to say "that's great but I'm married" and I was in a headspace where if I wanted to I would have gone with her for a run regardless of what my wife thought about it, but I also got the feeling this young woman will not take it well if I said I was married, so I just started avoiding her. I noticed she tried even harder to approach me. Later that evening one of of the women in the group asked me "Oh wow, so I checked your facebook and you're married?". And I said, "Yes." . That young woman just blurted it out "But I asked you to go running together, why aren't you wearing your wedding band????".
The point I'm trying to make is I'm sure having a great physique is sure to be helpful even in your 30s and 40s and it will certainly set you aside, but there are other, more powerful tools to build attraction, alas, some quite manipulative (like social proofing, oblique approach, status, feigned low interest, push pull, triangulation).
I think it's very important to understand where you stand, where your target demographic stands, and where your competition stands.
If you're in an area where all the guys are super fit millionaires, you're not going to have much luck with the ladies there.
I was never in the pickup scene, and I've seen some terribly messed in the head men because of it (bitter redpill types), but I think there are quite a bit of spoken or not so spoken psychoevolutionary tricks out there great to know and take with a huge grain of salt.
Probably the biggest trick is no trick, but to actually be ok with yourself, empathetic, reasonable, have fun by yourself, explore, have healthy boundaries, but still generous, not have/display emotional baggage, basically present yourself as being what the kind of woman you want also wants.
The thing that motivated me to start was the fact I was not very successful with girls and gaining 30 pounds of muscles in early 2000's Romania was intoxicating, if anybody told me before that girls would send kisses in the subway, grab my arms in the bus and start conversations with me or ask for my number in clubs, colleagues ask me to dump my girlfriend I would have said it's impossible.
I'm ashamed to say, but all that validation was even better than the way lifting made me feel and the primary drive to weight lifting.
It's only now that I remember how good weight lifting in itself made me feel, I never did give it much thought back then.
But now it's very hard to find the time or motivation to start it again.
I'm not really scared of getting fat down the line, I'm in my early 40s now and I've never been fat.
You could be right, that it's only the change that makes you feel amazing, and I only ever went to the gym for some 6 months total, but I have my doubt that it would ever go away, I've been on many, many drugs, NOT ONE ever made me feel good for 6 months straight, they all downregulate very fast.
Now thinking about it, I get a renewed motivation to re-start weight lifting
Completely anecdotal, but when I was 18, in highschool, I trained in the gym in my hometown, supervised with a trainer, 12 reps per muscle group, very modest gains.
I move to university, start reading a fitness forum where people were saying do max 6 reps if you want big gains.
I also started supplementing with whey protein, and within 3 months the gains were spectacular, everybody noticed, I felt on fire, best time of my life, I miss so much how great I felt in my own body.
I've seen other colleagues and how they trained -- I can say there was 100% correlation that those people who were not training hard also did not have big gains. People who had enough breath left in them to chat in the gym simply did not gain as much as people I saw as training hard.
Also for me, the 6 reps to exhaustion felt completely different then 12 reps (again, to complete exhaustion) -- immediately after the training it felt amazing to be alive, the world became a comfortable place, my anxiety completely vanished, and in the night and morning after an intense training (especially the legs and back) the erections and libido boost were out of this world, something I never felt with the 12 reps regimen.
Does that really work, though? I think it doesn't -- think all the anti-vaccine type influencers -- their identities are known and they're ok with it.
> It means creating a trusted identity that can be verified, and that the identity is known to be a real human with a reputation to lose if exposed as being a bot or otherwise untrustworthy.
Surely this won't be used for nefarious reasons or to silence individuals like it's done in the UK or in the cancel culture actions. /s
Nobody forced the Soviet Union into anything. I think the soviet leaders knew that the system in the West AT THAT TIME was simply better in all ways imaginable and the comunism utterly failed at its mission -- the workers in the West were enjoying a much much better life that those in communism, and having lost that ideological space, they thought they could override common sense on the battle field -- surely, if you win the space race, more olympic gold medals or on the battlefield, then communism actually won?
First, China is nothing like the USSR economically and the West is NOTING like the old capitalist West in any regard. Second, the ideological capitalism of the West during the Cold War is not what actually brought prosperity to the masses, I think it was just the fear of comunism that kept the elites at bay and willing to give some scraps to the unwashed masses.
I have ADHD and I also have hyperfocus, I think hyperfocus is an advantage in a pre-industrialized world.
As a child I was fascinated with blowguns. After a summer of shooting unripe grapes out of plastic pipe, I could shoot anybody in the forehead from 20 meters away, easily. I shot the blowgun thousands of times a day, it was relentless.
The same when I went fishing, a whole day could vanish and it would feel like a blink of an eye.
I taught myself how to ride a bike and I woke up that night to ride the bike, even though it had a flat tire.
I like to go mushroom hunting, but when I do, I usually like to go alone, I walk for extraordinary distances, rough terrain, I don't get bored, I can literally keep at it for the whole day that people think I'm crazy.
It's a bit like a stimulant induced obsession, but my inner voice recedes far back in my skull, it's an incredible flow state-like feeling.
I'm sure this kind of obsession builds skills and it has to have some benefits in pre-industrial societies.
The concept of envy/malice/insecurity and people lying to your face and stabbing you in the back was completely foreign to me up to the age of 36. Only in the face of overwhelming evidence and harm to myself did it all click.
Lately I'm seeing myself in this junior dev I'm mentoring, I'm strongly suspecting he's on the spectrum (that's why he was rejected initially from an internship, despite my input that he'd make a great dev, which proved 100% accurate) -- the guy is totally happy in his technical world, jabs and callous remarks from others completely go over his head.
A lot of people on the spectrum simply have a deep interest in things and systems. I could be wrong, but I think some of those spindle neurons and circuitry made to model others just get used in some people to get systems.
I often get frustrated because people seem to want to learn HOW a technical insight and it's impossible for me to tell them HOW I got to that conclusion, other than I deeply immersed myself in it and it just clicks. I get the same awe when my wife makes jokes about a behavior of mine or someone else and I can see just how deep, funny and plausible her whole internal model of others is; and sometimes how wrong it is, just like my internal model of a system sometimes is. Alas I can change my internal models of systems on a whim.
It's not legal where I live also, I did try 2-FMA and it felt better in certain scenarios -- like following a hard course, but I also felt the tolerance ramps up much faster in releasers than re-uptake inhibitors so methylphenidate still is a wonderful tool.
My take would be that coding itself is hard, but I'm a software engineer myself so I'm biased.
I want to share my counter example -- no amount of therapy could help me not almost loose it every time I drove my daughter in heavy traffic or deal with her just being a toddler.
But after 2 weeks being on concerta made lasting changes even months off the drug.
It was the best type o therapy -- you just do the thing that triggers you minus the bad part and learn it's not so bad and you can do it, it has profound implications.
I have ADHD and take metylphenidate(I've tried many kinds of stimulants as well) -- and the NO2 analogy is an imperfect but better analogy than saying stimulants simply adjusts the baseline of people with ADHD to function like "normal" persons.
I feel there is a narrow window of dosage and time where it might feel that way -- i.e. stimulants at the onset might calm you down, reduce anxiety, but all stimulants are very broad hammers.
For me it feels like it's impossible to re-create chemically exactly the neurotypical focus that I've seen in other colleagues.
Like spending 5-6 hours of continous work where you drill down just enough, get back on track, don't get distracted, don't get too anxious, don't get hyperfocused AND do that consistently, day after day after day.
My non-chemical modes are either hyper focus for 2 weeks on a problem, immerse myself but then completely lose interest, most of the time without showing much for it OR procastinate it a long way, get extremely anxious and work really hard on the problem.
With stimulants it's a bit like: - dosed just right:it evaporates anxiety, stressful situations feel easy to deal with, BUT there's always increased heart rate, grinding teeth and some tension at the end of the day - some stimulants make mundane things wildly interesting (on isopropylphenidate I spent a few hours playing with a PLSQL debugger because I thought it was really cool), but no sense of "GO, GO, GO, do it". - some make things seem urgent enough and help stay on track -- like the metylphenidate I'm prescribed. - some make going into a flow-like state easy and fun (like methamphetamine and phenmetrazine). - some are pure energy and urgency -- like modafinil.
All of the stimulants have the potential to give me euphoria, all of them temporarily increase libido I still have to be mindful of not focusing on the wrong thing, the "normal" feeling is very fleeting, it's very easy to get hyper on stimulants, all of them feel like wear & tear at the end of the day, some more than others.
Genuine question out curiosity (I have a master in finance, but never practiced it) -- aren't both the cloud bill and depreciation all tax deductible, eventually? the bill 100% in that year and the depreciation spread over multiple years?
> Hybrid has been the answer to both. It shouldn't be a binary decision. stateless compute workload can fairly easily be offloaded to private cloud.
Can you elaborate on that? I'm studying for saa-c03, and I was shocked by how expensive egress out of aws can be.
I'm a java dev and maybe my projects are about big integrations, but I've always needed queue like constructs and polling from a db was almost always a headache, especially with multiple consumers and publishers.
Sure it can be done, and in many projects we do have cron-jobs on different pods -- not a global k8s cron-job, but legacy cron jobs and it works fine.
Kafka does not YET support real queue (but I'm sure there's a high profile KIP to have true queue like behavior, per consumer group, with individual commits), and does not support server side filtering.
But consumer groups and partitions have been such a blessing for me, it's very hard to overstate how useful they are with managing stateful apps.
Maybe your intentions are good, but remember, unless we legalize polygamy, the "bad/inept/creepy straight-white men" narative should crumble for people in their 30s and 40s when it's the last train for marriage and children.
But we don't have "some of those women..." narrative about single women in their 40s complaining they can't find a husband.
My point is it's an universal problem in the civilized world, spanning vastly different cultures in Asia and Europe and North America, "some of those men" is very hand wavy explanation, and I think it spans from the extremely toxic (I'd say anti-human and demonic) hollywood pop culture.