o10449366

No wonder. It's performance overall was noticeably, like it had regressed to coding models from 1.5 years ago. I've try not to use claude during peak US hours because it tends to struggle more then with reasoning and correctness it seems than off hours.

Can you tell us more about the motivation for this project? I'm very curious if it was driven by a specific use case.

I know there are specialized trading firms that have implemented projects like this, but most industry workflows I know of still involve data pipelines with scientists doing intermediate data transformations before they feed them into these models. Even the c-backed libraries like numpy/pandas still explicitly depend on the cpython API and can't be compiled away, and this data feed step tends to be the bottleneck in my experience.

That isn't to say this isn't a worthy project - I've explored similar initiatives myself - but my conclusion was that unless your data source is pre-configured to feed directly into your specific model without any intermediate transformation steps, optimizing the inference time has marginal benefit in the overall pipeline. I lament this as an engineer that loves making things go fast but has to work with scientists that love the convenience of jupyter notebooks and the APIs of numpy/pandas.

Comments on AI, on an article which has nothing to do with it, are just as uninspired as AI writing.

I mean, you're making assumptions about the author's intent going one way, but not the other. What if the polite tone is what they intended? And how do you know they didn't review the output for phrasing and fabrications?

The author acknowledged they used AI to translate. Is the translation they decided to publish among the given tools they had available to them not by definition the most authentic and intentional piece that exists?

All of this aside, how do you think tools like Google Translate even work? Language isn't a lookup table with a 1:1 mapping. Even these other translation tools that are being suggested still incorporate AI. Should the author manually look up words in dictionaries and translate word by word, when dictionaries themselves are notoriously politicized and policed, too?

When the translation differs so much from what the author is trying to say in their native language, it loses its earnestness.

That's why translation is a job in the first place and you don't see publishers running whole books through Google translate. No one, least the authors, would accept that.

Highly doubt this. Have you read a translated book? Are you looking for literal translations or a translation from someone who's an expert in both languages and makes subjective adjustments based on their experience?

Why? Would a incorrect but literal translation be closer or further from what the author is trying to communicate?

I've been seeing this take on HN a lot recently, but when it comes to translation current AI is far, far superior to what we had previously with Google Translate, etc.

If the substack was written in broken English there's no way it would even be appearing on the front page here, even less so if it was written in Chinese.

That's not true from a mechanical perspective. Most SUVs use the same frame and parts as trucks by the same manufacturer (which is why they handle so poorly compared to sedans - it isn't just center of gravity)

If uv figures out a way to capture the scientific community by adding support for conda-forge that'll be the killshot for other similar projects, imo. Pixi is too half-baked currently and suffers from some questionable design decisions.

What would an equivalent tool be on linux? I guess it depends on the filesystem?

I used to work at amazon and had a medical exception for working from home. While obtaining the exception, the HR person in charge of my case would repeatedly call my personal cell phone to ask me questions about my disability. They did this 4-5 times despite my insistence that we keep all correspondence written and over email and despite me fulfilling all listed documentation requirements. Once my exception for my chronic condition was approved, they noted that I would need to renew every 6 months, because I guess lifelong conditions you're born with warrant constant validation.

maybe someone will finally highlight how ridiculous the gridlock is on the b44-sbs route, particularly through south williamsburg. I regularly see convoys of 4-5 buses arriving at the same time because the traffic through that neighborhood is so bad that the buses eventually catch up to each other and I regularly have to wait 30+ minutes for it on either end of the route.

Crazy that you can't find any 7soulsdeep graffiti on here

My hair is so course and thick that I can only shave with feather blades, but I hate how every time I travel I can't bring them with me.

I bought disposable safety razors that are single blade and marketed in India and while they are better than all of the multi blade disposables, they still suck. Does anyone have a solution for good shaving while traveling if you can't bring your DE razor?

and just like that - gone off the front page

This is already a feature and has been for a long time. Instagram allows you to only see posts from people you follow and will tell you when you've seen all new content from the past X days.

Yeah, I just went through two onsites with Kensho (for two different teams). After the first onsite, I asked them if their compensation range would be competitive with another offer I already received and the recruiter told me "they are very similar." I spent another 6 hours doing the second onsite only to be told that their best offer is 35% lower due to "budget cuts." Needless to say I'll never be interviewing with them again because they've shown they don't respect my (or their interviewer's) time.

This is a great idea. I can only speak to NYC, but the data sources don't give good recommendations. Optimal routing for time in NYC is tricky though and relying on only mapping services and not a combination of walking, biking, buses, and subways leads to longer travel times than necessary.

For example, there are many parts of Brooklyn where no transit lines are directly connected and you have to go into Manhattan first, meaning if you took the train it would take 1+ hours, versus bus (30 minutes) or biking (15 minutes).

I've had the same experience at the big tech company I worked at.

I expressed these frustrations to upper management and the explanation I was given was that for many of these big companies, their goal is to grow their 1,000 sub-businesses to 10 million ARR businesses each. Growing these 10 million ARR units to 100 million ARR is much more difficult than going from 0 to 10 million, and so when they have a successful unit, they disproportionally reinvest that success into the poor performers (the 0-2 million ARR units).

This felt backwards and unfair to me as an engineer on the teams that worked hard to get to the 10 million mark. It really sucked seeing my team lose headcount or engineers to less successful and poorly designed/managed projects. But from a business perspective on the larger scale I see why they do it - I just think it creates a poor engineering culture and, in the long-term, a sub-par customer experience because velocity in product development slows when projects are de-invested or put into KTLO.

Customer-friendly? Maybe if you've never required support or human engagement for any issue. If you have, Google is by far the worst among the big tech companies.

The article is right that intentionality in relationships is critical. I think it's a commonly romanticized view that relationships shouldn't feel like "work" if they're good relationships, but I think that view does a disservice to the effort that goes into building a deep and fulfilling connection. Lifelong connections don't happen by accident.

I also agree with csallen's comment. Introducing my friends to each other was one of my projects for last year and its made my whole world closer and more convenient. Now if I ever want to see a movie, have a day in the park, have a dinner I know I can make a general invitation and whoever is free and interested can show up and, because they already know each other, everyone will have a great time.

If you trust your friends to see in each other (or at least respect) what you see in each of them individually and if they're mature adults, there should be no issue with introducing people to each other. This practice has been a good way for me to re-examine some of my own unconscious biases about which of my friends would get along with one another based on superficial qualities (such as their occupations or backgrounds) and I've been pleasantly surprised by the friendships that have blossomed out of the introductions I've made.

As a side note, I still identify as an introvert and still get the most fulfillment out of 1:1 conversations. I don't think you need to be an extrovert to hang out in groups or be the social glue in your world. I was inspired to start introducing friends to one another because I realized how grateful I was - as an introvert - when I first moved to a new city and one of the friends I made was super welcoming and inclusive with their friends. It made me want to serve in that role myself and I'm so glad I got over my shyness and anxiety. I think if you approach adult relationships with intentionality you'll often be surprised by how receptive people are - I find most people have a desire to make new friends, but they're often afraid to take the first initiative.

just goes to show how confused Amazon's product strategy is when these models have already been available on JumpStart. When a company becomes large enough that teams start developing parallel products without collaboration, that's when you know its grown past the point of efficiency.

My wife, bless her heart, leaves everything partially open. Not even just open, but she'll screw the lid on jars just enough where if I pick it up by the top they'll drop. It drives me crazy.

You're generally right, but I think the strong dependency on GPUs with specific architectures and capabilities and how scarce they are is a fairly unique problem to ML (outside of crypto mining lol.)

Sure, MLOps is just Data Engineering in disguise when you ignore the complexities of hardware provisioning, GPU optimization, integration tests for model performance and quality, benchmarking, resource constraints (network, disk, memory, GPU memory), etc.

Anecdotally, I've worked in high-performance computing and machine learning for years now and the past few months I've seen a huge spike in the number of messages I get for MLOps positions. I think companies are slowly starting to realize that setting up machine learning at scale isn't as simple as deploying poorly written code by research scientists to managed platforms.

If I get approved for a remote location I get a huge paycut, despite continuing to pay the same taxes and COL in the same location.

Drugs like Adderall don't mimic the hormones talked about in the article, right?

I took Adderall XR once a day in college which completely suppressed my appetite and, combined with being a poor student that couldn't always afford lunch, I would frequently only eat one meal a day (either breakfast or dinner.) Since I graduated college many years ago, I also stopped taking Adderall but my sense of hunger has never returned. I always figured my body just got used to having my appetite suppressed and now I can go multiple days without eating and not feel any physical sensation or side-effect (including fatigue/irritation from lack of food or things like headaches.) Just the other day I went 40 hours without eating while traveling for the holidays and didn't even realize it until I went to a social dinner.

I've never thought too much of this. It seems like in pop culture/science nowadays "fasting" is generally seen/understood to be beneficial. I certainly don't mind because its helped me stay at a healthy weight, but in reading about this new class of drugs for weight loss recently I am curious as to what exactly is going on in my body and whether there's a similar mechanism at play.

Is there anyone who fasts regularly who's also experienced this loss of hunger sensation?

Unfortunately, it's a pretty tired HN comment to posit something as being more simple than it actually is.

Even during the pandemic there were widespread complaints about the signal work on the L line that has now transformed it from one of the worst lines to one of the best. There's no such thing as "simple work" on a system that millions of people depend on for consistent uptime.

Honestly, the voting sorts on Reddit are an underrated feature. Being able to sort comments not by total number of upvotes (which I feel panders to low-effort content) but by highest ratio of upvotes to downvotes produces an extremely different experience. I also enjoy being able to see "controversial" comments because sometimes those can be quite insightful.