HNHacker News
TopNewBestAskShowJobs

beoberha

1,053 karma · joined February 15, 2020

submissionscomments
beoberha··on Databricks in talks to acquire startup Neon for about $1B
Congrats to the Neon team - they make an awesome product. That’s about all the good I can say here. I don’t blame them for selling out. It’s always felt like a “when” not an “if”. I would be surprised if you can make money selling cloud databases - especially when funded by VCs.
beoberha··on ClickHouse gets lazier and faster: Introducing lazy materialization
look at who you’re arguing with ;)
beoberha··on ChatGPT now performs well at GeoGuesser
Assuming it only used the pixels and not any metadata from the file or memory from the user (which is a massive assumption), how fucking cool that it can identify the Olympics versus any other mountain range. At that point it’s probably not too hard to guess the picture came from Alki or Golden Gardens, but still very impressive!

I’m also completely ignoring it inferred location from the pride flag and corgi which have heavy Seattle vibes :)

beoberha··on OpenAI looked at buying Cursor creator before turning to Windsurf
People clearly want the subscription model where they don’t have to worry about API keys and such. I bet a huge chunk of this market is non technical people who can’t code and don’t realize how bad the code they’re writing is when Windsurf and Cursor chop off the context to make it cheaper to run.
beoberha··on Distributed Systems Programming Has Stalled
Yep - I’ve very much been living the former for almost a decade now. It is especially difficult when the components stretch across organizations. It doesn’t quite address what the author here is getting at, but it does make me believe that this new programming model will come from academia and not industry.
beoberha··on Show HN: LLM plays Pokémon (open sourced)
Same! Except I tried using LLaVa 1.5 locally and it didn’t work at all lol
beoberha··on MongoDB acquires Voyage AI
I think the amount of people working on large enterprise systems here is a lot smaller than one would think.

Whenever a fly.io post about sqlite ends up in here, there are a scary amount of comments about using sqlite in way more scenarios than it should be.

beoberha··on You're not a senior engineer until you've worked on a legacy project (2023)
Oh man this one hits home. I don’t do much coding anymore but my general advice to folks I lead is you’re never going to be happy with how you did things and just make sure it scales and is well tested.

Edit: oh and how could I forget as simple and readable as possible

beoberha··on Basketball has evolved into a game of calculated decision-making
I don’t think this is a particularly well written article, but I sort of agree with the sentiment. Basketball just isn’t THAT complex and the talent pool is homogenous enough that most teams can find these archetypes and build rosters that get you to the playoffs.

That said, trends are cyclical. Look at the role of the running back in the NFL. There will always be outlier players like Shaq who will buck the trends and exploit matchups.

beoberha··on SQL pipe syntax available in public preview in BigQuery
Eh, I don’t need Lisp. I need to do my job. KQL allows me to write queries intuitively and others to easily understand what I’m doing.

The summarize operator is also a game changer. I use something like “ | summarize count() by bin(TIMESTAMP, 1h) “ multiple times daily. Getting the same result in SQL is disgusting.

beoberha··on SQL pipe syntax available in public preview in BigQuery
Somehow Microsoft (my employer) has totally failed in evangelizing Kusto to the masses. KQL is by far the best query language you’ll ever use.

This syntax from Google is nice but it’s still just SQL.

beoberha··on Launch HN: Karsa (YC W25) – Buy and save stablecoins internationally
Not true - swapping is a taxable event. Doesn’t matter which coins
beoberha··on Ask HN: Industry Wide Strike for Worker Protection?
It’s completely infeasible, but I wouldn’t be surprised to see significant movements inside individual companies or small groups of large companies.
beoberha··on Advent of Code 2024 in pure SQL
Thomas Neumann doing Thomas Neumann things
beoberha··on Things we learned about LLMs in 2024
LLMs being able to detect bugs in my own code is absolutely mind blowing to me. These things are “just” predicting the next token, but somehow are able to take in code that has never been written before and somehow understand it and find what’s wrong with it.

I think I’m more amazed by them because I know how they work. They shouldn’t be able to do this, but the fact that they can is absolutely jaw dropping science fiction shit.

beoberha··on OnlyFans models are using AI impersonators to keep up with their DMs
Someone in the “side projects making $500+ per month” thread yesterday said they have a business doing this
beoberha··on What made Dostoevsky's work immortal
My favorite course I took in college was a Russian Literature elective my first semester. I fell in love with Dostoevsky when we read Crime and Punishment and I ended up writing my 25 page term paper on him and his role in the proto-existentialist vs nihilism movement in Russia in the 1860s. There will always be a part of me that wishes I had taken my career in that direction instead of computers :)
beoberha··on Meta Uses LLMs to Improve Incident Response
You’re better off reading the original post instead of this ad: https://news.ycombinator.com/item?id=41326039

On the general topic - I’m interested in people sharing their experiences using LLMs in livesite scenarios.

beoberha··on BERTs Are Generative In-Context Learners
Wanted to circle back. The 2.5 coder 7b instruct is amazing, holy crap. It classifies my cases almost perfectly and almost always returns perfectly formatted JSON.
beoberha··on Beating the bookies with their own numbers
Anyone who makes a living off of sports betting is paying other people to put the bets down for them. Even the in person casinos will ban you if you win too much.
beoberha··on BERTs Are Generative In-Context Learners
Depends what you’re trying to do. I’m writing a personal assistant app (speech to text) and want to classify the user input according to the current actions I support (or don’t). The flagship LLMs are pretty great at it if you include the classes in the prompt and they will spit out structured output every time. But, man, they are expensive and there’s the privacy aspect I’d prefer to adhere to. I’ve only got 24 GB of RAM, so I can’t run too many fancy local models and things like llama3.1:8b don’t classify very well.

So I’m trying BERT models out :)

beoberha··on WebSockets cost us $1M on our AWS bill
Classic Hacker News getting hung up on the narrative framing. It’s a cool investigation! Nice work guys!
beoberha··on Pushing the frontiers of audio generation
Totally agree. Maybe it’s just the clips they chose, but it feels overfit on the weird conversational elements that make it impressive? Like the “oh yeahs” from the other person when someone is speaking. It is cool to see that natural flow in a conversation generated by a model, but there’s waaaay too much of it in these examples to sound natural.

And I say all that completely slackjawed that this is possible.

beoberha··on Adventures in algorithmic trading on the Runescape Grand Exchange
Thanks! But how was the bot running before the training data existed? I think that’s what I was missing. The baseline?
beoberha··on Adventures in algorithmic trading on the Runescape Grand Exchange
Sorry if I’m just bad at reading comprehension, but what exactly is being modeled here? It sounds like the author is just getting price data not unlike a candlestick chart. What I don’t understand is if the trading is happening within the 5 minute windows he mentions or if model is finding patterns across timeframes?
beoberha··on Quantized Llama models with increased speed and a reduced memory footprint
For me, it was almost random if I would get a little spiel at the beginning of my response - even on the unquantized 8b instruct. Since ollama doesn’t support grammars, I was trying to get it to work where I had a prompt that summarized an article and extracted and classified certain information that I requested. Then I had another prompt that would digest the summary and spit out a structured JSON output. It was much better than trying to do it in one prompt, but still far too random even with temperature at 0. Sometimes the first prompt misclassified things. Sometimes the second prompt would include a “here’s your structured output”.

And Claude did everything perfectly ;)

beoberha··on Extracting financial disclosure and police reports with OpenAI Structured Output
Stuff like this shows how much better the commercial models are than local models. I’ve been playing around with fairly simple structured information extraction from news articles and fail to get any kind of consistent behavior from llama3.1:8b. Claude and chatGPT do exactly what I want without fail.
beoberha··on AI agents invade observability: snake oil or the future of SRE?
I agree completely. SRE/Ops/Livesite is an incredibly hard problem and very easy to make shiny demos for products that will not reproduce those results when you need them most.

The article talks about moving past “copilots” and moving right to “agents”. There’s probably some semantics to decipher there, but we haven’t even gotten copilots to work well! At their core, they are essentially the same problem, but I feel a lot safer with a chatbot suggesting a mitigation than just going and performing it.

beoberha··on Amazon tells employees to return to office five days a week
This is exactly it. The few years I had in office were an amazing foundation for my career. I see the same in those who started around the same time I did. Most of my team who was hired remotely are struggling.
beoberha··on ReMarkable Paper Pro
Woah I’m shocked! I’ll definitely be doing that.
← PreviousPage 3 of 9Next →