HN user
Interested in AI, Blockchain, infrastructure, data stuff.
Corlinp@ the Google mail service
Wow- the entire Point Loma area pretty much looks the same as it does today. Too bad they didn't drive down Newport to see Hodad's.
Torrey Pines area definitely looks the most different, mostly because of the growth of UCSD I'm thinking.
That one is a bit sus to me, because the models that do the worst on Omniscience Accuracy do the best on non-hallucination. The top model for this benchmark is "MiniCPM5-1B (Non-reasoning)" which gets a whopping 99% vs 45% for Fable 5.
I'd love to see a good hallucination benchmark, but this isn't one. There's no possibility that a 1B model hallucinates less than Fable 5.
Most of those inference providers are American, and China is actually at a disadvantage here because of export restrictions - US companies are using newer and more efficient chips.
This should erase any doubt that AI Labs are making $$$ on API inference.
Kimi 2.5 (which this is based on) is served at $0.44 input / $2 output by a ton of different providers on OpenRouter, 2.6 will certainly be similar.
That's about 11X less than Opus for similar smarts.
I'm running it for the first time and this is what the thinking looks like. Opus seems highly concerned about whether or not I'm asking it to develop malware.
This is _, not malware. Continuing the brainstorming process.
Not malware — standard _ code. Continuing exploration.
Not malware. Let me check front-end components for _.
Not malware. Checking validation code and _.
Not malware.
Not malware.
IAM is generally binary (allowed yes/no) whereas MCP usually supports more nuance (always allow/ask if risky/always ask/no)
You'll have to get back to us with that name!
Well I suppose I can't miss out on the opportunity to plug my open-source menu bar app for voice-to-text in any app! Going on three years of development, believe it or not.
When I set up a Mac I have a short list of things that I need to install. When I set up Windows I have a much longer list of things that I need to un-install. I much prefer the former.
iStat menus (or the new open source "Stats") is a brilliant use of the menu bar, but it can take a lot of space!
Then couple ordinary services that add menu bar icons that you don't even ask for (DropBox, Docker, Adobe, etc.) and you can overflow onto the notch quite quickly.
Every time I get a new Mac, I run these commands to reduce the spacing between menu bar icons. Lets you fit at least 2x the number of items in the menu bar.
```
defaults -currentHost write -globalDomain NSStatusItemSpacing -int 2
defaults -currentHost write -globalDomain NSStatusItemSelectionPadding -int 2
```
This is exactly the case today. Multimodal LLMs like gpt-4o-transcribe are way better than traditional ASR, not only because of deeper understanding but because of the ability to actually prompt it with your company's specific terminology, org chart, etc.
For example, if the prompt includes that Caitlin is an accountant and Kaitlyn is an engineer, if you transcribe "Tell Kaitlyn to review my PR" it will know who you're referring to. That's something WER doesn't really capture.
BTW, I built an open-source Mac tool for using gpt-4o-transcribe with an OpenAI API key and custom prompts: https://github.com/corlinp/voibe
I created Voibe which takes a slightly different direction and uses gpt-4o-transcribe with a configurable custom prompt to achieve maximum accuracy (much better than Whisper). Requires your own OpenAI API key.
https://github.com/corlinp/voibe
I do see the name has since been taken by a paid service... shame.
Guessix is an LLM-powered word game like guess who!
We have a fun group working on it on Discord (find the discord invite in the How To)
Very interesting! Do the benchmarks hold up well or does it reduce performance in other areas too?
Do you mean like structured outputs? Unfortunately here the model is guided to explicitly tell you when you violate the rules and why, it can confuse it's system rules with the game rules and say you're not allowed to ask a question about copyrighted material etc.
The performance is great, but the censorship is ridiculous for me. I tried it as a backend for my game Guessix[1], but it would refuse for ridiculous reasons like "Cannot answer questions about copyrighted works like Harry Potter."
Amazon has a number of foundation models under the name Amazon Nova, which they claimed were SOTA on release but I haven't heard much at all about them since.
ParaFi Capital | San Diego (Hybrid) | Sr. full-stack Engineer
ParaFi is a multi‑strategy investment and technology firm focused on the digital‑asset ecosystem. Since 2018 we’ve partnered with leading institutions to deploy venture, liquid, and quantitative capital. Our edge comes from marrying rigorous research, deep technical capability, and an operator’s mindset. We ship code, contribute to protocols, and build internal systems that keep us a step ahead of the market.
We’re looking for a senior engineer who can wear multiple hats - backend, frontend, data, and DevOps - to accelerate every part of ParaFi’s technology platform. You’ll design and own production systems that ingest billions of on‑chain and market datapoints, power quantitative strategies, and surface realtime insights to the investment team. Expect a high‑trust, low‑bureaucracy environment where your technical decisions directly influence portfolio returns.
To apply, send me an email directly and mention HN in the title. Just decode this email: iuxrot.v@vgxglo.ius
Check out Voibe: https://github.com/corlinp/voibe
Open source Mac-native menu bar app for speech to text using GPT-4o-transcribe (current STT SOTA)
I used Suno to generate rap versions of Feynman's first lecture. It turned out really well!
https://suno.com/playlist/d2886382-bcb9-4d6d-8d7a-78625adcbe...
I've tried this with a few models in the past, and Suno v4.5 is the first to produce decent results! Enjoy the educational rap.
One thing I'm really excited for is having this real-time voice model in video game characters. It would be really cool to be able to have conversations with NPCs, and actually have to pick their brain for information about a quest or something.
ParaFi Capital | Remote | full-time | Full Stack Engineer
ParaFi was among the earliest investors in the blockchain industry and has evolved into a trusted partner by leading institutions globally. At ParaFi, you'll join a cross-disciplinary team united by collaboration, excellence, and optimism for the future of the internet.
We're seeking a talented full-stack engineer, prefer experience in React and Go. You'll build data-driven dashboards and analysis to support our portfolio and risk monitoring, trade systems, and more.
You can apply by sending a thoughtful email to careers@parafi.com
ParaFi Capital | New York City | full-time | IT Systems and Security Administrator
ParaFi was among the earliest investors in the blockchain industry and has evolved into a trusted partner by leading institutions globally. At ParaFi, you'll join a cross-disciplinary team united by collaboration, excellence, and optimism for the future of the internet.
We're looking a passionate sysadmin to support and protect our firm's digital ecosystem. The position offers an opportunity to work at the intersection of cutting-edge finance and advanced cybersecurity for the evolving Web3 threat landscape.
https://docs.google.com/document/d/1nyCMHDk97efhU1VLFYkjT1ou...
You can apply by sending a thoughtful email to careers@parafi.com
ParaFi Technologies LLC | U.S. remote possible, San Diego preferred | full-time | Infrastructure Engineer
ParaFi Technologies is a small team operating some of the highest-performing blockchain validators and node infrastructure. We are looking for a passionate, junior- or mid-level engineer to develop software and infrastructure (Python, Go, Terraform, Docker, etc.) and be an integral part of ParaFi Technologies' growth.
https://docs.google.com/document/d/e/2PACX-1vQUBpZCVa8fmfJiO...
You can apply by sending a thoughtful email to careers@parafi.tech
ParaFi Technologies LLC | U.S. remote possible, San Diego preferred | full-time | Infrastructure Engineer
ParaFi Technologies operates some of the highest-performing validators and node infrastructure for blockchain networks.
We are looking for a driven, junior- or mid-level engineer who can develop software and infrastructure (Python, Go, Terraform, Docker, etc.) and be an integral part of ParaFi Technologies' growth.
Details: https://docs.google.com/document/d/e/2PACX-1vQUBpZCVa8fmfJiO...
You can apply by sending a thoughtful email to careers@parafi.tech
Because you're looking forward at the platform that's moving down and toward you. Kind of like if you were to stare down at the belt of a treadmill while it's moving - it would be disorienting after a while.
Finally I get to be that guy in the comments with a weird about of relevant experience. I built the world's largest hamster wheel in 2012[1], a large rotating circular platform ~6m in diameter.
It was a fun and unique experience to run on for a short amount of time, but most people would get dizzy after a few minutes of jogging on it. The curved platform also turned out to be a bit of a tripping hazard. It was more often used as a sort of swing (could this work on the moon?).
I'm skeptical that the experience on the moon will be much better, especially since the diameter they're proposing is even smaller.
1. https://sdusd-newsfeed.blogspot.com/2012/09/pt-loma-high-sen...
ParaFi Technologies | Blockchain Infrastructure Engineer | US Remote (+San Diego based) | Full-time | https://parafi.tech/
We're a small team of engineers running some of the highest-performing blockchain validators and node infrastructure. Looking for a passionate engineer with ~2-6 YoE who wants to both develop software + infrastructure (Python, Go, Terraform, Docker...), and participate in other parts of the business, from financial modeling to talking with customers. Get the fun of a startup with the stability of an established investment firm!
For more info, email [my username with a dot before p]@parafi.tech