wkoszek

Hey. I liked your feedback, so I updated the website.

You can give it a shot in VM - I've done testing on macOS, and then used Ubuntu 24.04 home box to install it, also in multipass. In GitHub I put howto.

Sounds good. bsub.io will offer a way to be self-hosted 100% offline too. Just not now. That will require some work.

It's not self-hostable yet: I'm thinking about adding this option. I think for now I'm trying to see if people would like to give it a shot and try it out.

Why do you think it's a deal breaker for you?

US is the same. I no longer know is someone is using long-distance lights or normal lights, or whether they are wrongly adjusted. But driving at knight, especially with folks right behind you is hard.

This is a great feedback! You can play with it some more if you Login. I have put a playground for users who'd like to try it out without installing.

But it's a good point--perhaps I need to add more examples to the page so that you can see right away what it's all about.

Thanks! Perhaps you can test new upcoming workers. There'll be more tools for document processing and rendering, with some examples of what can be done with them.

Thanks! Any feedback is appreciated!

The effect is beautiful. Is there a way to easily get the very same effect in TeX or some other text -> PDF format?

https://www.bsub.io - batch processing for developers.

You submit heavy duty jobs without worrying about infra, and we take care of execution. We're starting with PDF extraction. Audio transcoding + STT (speech to text) is next. Video transcoding will follow.

This allows you to have $5/mo VPS and get media operations figured out.

It's interesting how nobody talks about due-diligence being completely broken. We raised $$$ from many VCs and the DD for some of them was crazy: line item by line item with calls to customers etc. Tech folks were on phone with me and had to explain them stuff step by step, revealing a lot of confidential recipes. Also did this for bigger customers. And the $175M deal.. isn't there an earnout? Like $10M cash now, 1/4*$175 wired on 1yr cliff, and then the rest over 4 years if some milestones are hit? The whole thing looks weird.

From Claude -> Notable Programs They Fund: Public television shows like Sesame Street, NOVA, PBS NewsHour, and Masterpiece, as well as NPR programming like Morning Edition and All Things Considered.

I like all of those. NPR: $300m budget / 42m listeners = $7.14/yr. Sounds like if I donate $5/mo to KQED and $5/mo to KCSM, I'm supporting them to cover myself and couple other citizens?

I don't get what I can do to support PBS - when I press donate on PBS site, it sort of wants to direct me to KQED/KCSM donations again.

Anyone here with a little more time to understand/explain it?

Working on the notes app with MCP, JSON export and Git backup. Like Apple Notes, but works on every platform and in the browser. Offers focused mode where you can just full-screen the app and type your notes.

The big question now would be: has anyone used BoldVoice or any other method to convert from 90%+ of accent A to 90%+ to accent B and can switch between those seemlessly?

I don't use anything, but each time my eyes hurt, I wonder if I should: with some sense of music and tones, I often wonder if I could develop enough audio feedback to be able to do tasks like opening apps, writing, compiling, and getting feedback. I also often wonder if that is something that the visually impaired already use.

Great work! I really hope that you guys end up having a ton of industry interest. Really see JITX being a positive addition to the whole EDA toolchain. I remember several efforts from within the industry where even during copy&paste from reference designs, the mistakes were made and the PCB batch was ending in a trash.

Once the library of snippets gets beefy, and people can build boards with their Raspberry Pis and Zeros on them, it'll be a critical mass.

Good luck!

It should work on Mac Studio with Ollama

ScratchDB has save my business and it's awesome. I think if you need a columnar store, you should really try these guys

I regret missing the Open Source Firmware Conference--I really wanted to meet them and learn some more on what's the story here.

I know it's a brilliant team, and since ZFS/OpenSolaris/Dtrace I've known about Bryan. The product is a nicely looking product too. One thing I don't get is the target market. Who is it? Banks? Cloud vendors? Insurance companies? I think there's a part of the computer business that I don't get yet.

One guy in one of the previous HN threads here said that it's convenient to run 1 RFP for 1 box costing $1M-$2M vs running 30 RFPs for isolated components to put on the rack. Ok, I can see that. That's some argument.

But I know that Supermicro has rack assembly service. https://www.supermicro.com/en/products/rack and I'm sure Dells and HPEs have the same. Does it mean that it's impossible to call them and tell them: "Excuse me, I want 2 racks, 15 systems each, top of the line AMD CPU with RAM, NVMe storage, all maxed out, and whatever fastest Juniper switches you can find. Put me VMWare vFusion on it. Ship this thing to Infomart in Texas. I have $2M here with your name on it". They won't do this for me?

Its either this, or Oxide is going to e.g.: Bank of America, and along with RFP the bank is asking: "Show me complete supply chain records, including the source code to your boot loader, drivers, and any firmware that any component on this motherboard is running". And maybe that's ... impossible these days?

I'm the startup CTO, so not the consumer base, but after DHH started posting his thing about cloud exit, I decided to explore this, and I walked the floor of 3 DCs (in SV/TX). People who walk you around the DC don't appear to care about cables, fans or noise etc. If it's longer than 1hr on the floor, they'll put Airpods in and they're done. I've seen cages that look unified, pristine, with love and affection put into cable layout, and I've seen cages that look like a total mess. It's your cage--you only go there if things break.

My last guess is that maybe Oxide rack will end up being sort of what an Apple Macbook is among cheap HP/Acer laptops from Walmart. And it'll be a shiny toy of bold bearded IT dudes who work for GEICO IT by day, but by night they scavenge eBay for used server deals and get excited about the idea of running their own private rack in their basement. They can't afford it at home, but at work they'll want their own Oxide. If you're reading this, do know I know who you are.

It's sort of interesting to see it. I believed that Mangini would be their last drummer.

My aspirational goal is to be in a position to retire early and hack on Postgres full-time, it’s so interesting. It has networking, storage, data, algorithms etc.

C is less of a problem tbh. Postgres has good code style and its pretty consistent. The complexity of the internals is, and having a small community may infer the speed in which people help you.

That would be fine--basically any way of achieving it would be fine. As of now, in PostgreSQL's FTS, I don't think there's any way to do this fast enough to give it back to the user.

Hey guys. Congratulations - this is an exciting development. Can you show some benchmarks around showing the count of matches -- `select count() from table where text match is there`?

This was the top reason that made us (Segmed.ai) give up on PostgreSQL FTS -- our folks require a very exact count of matches for medical conditions that are present in 20M reports. And doing COUNT() in PostgreSQL was crazy, crazy slow. If your extension could do simple len(invertedindex[word]) that would already be a great improvement.

ELK has it immediately, but at a cost of being one more thing to maintain, and the whole Logstash thing is clunky. I'd love to use FTS inside of PostgreSQL.

Can someone skilled in the art tell me where can I learn more about generating those puzzles? Are puzzles on chess.com hand-crafted, or are they generated with an algorithm?

It's awesome that you replied to this thread, because I was thinking about writing a program like this, and I'm happy that it exists.

Does it work on the modern OSes?

Your role as a company is to maximize revenue. If presence of bugs makes you lose revenue by customer churning or burned-out developers leaving, fix bugs. If not, create new features. And define a bug as "thing that the customer complained about". If customers don't complain, it's either not a bug, or you are solving a wrong problem.

Segmed (YC W20) | Engineers etc. | Full-time| REMOTE | https://www.segmed.ai/careers

To aggregate diverse data, we have partnered with many US-based and international hospitals and imaging clinics - about 600 healthcare institutions spanning 5 continents, to be exact.

We come, we integrate our healthcare ETL pipeline with a healthcare institution, and start de-identifying the institution’s data. We can do this for both historical and real-time data, as well as both text and images. We handle de-identification in a way ensuring that protected health information - or PHI - never leaves the premises of the institution.

Once this de-identified data transfer is established, we can aggregate our partners’ data inside of our Segmed Insight platform in the cloud. Insight is a web-app that our customers can enroll onto.

In Insight, users can search the data, and build patient cohorts appropriate for your research and development. For example, if you want to find chest CT scans of patients with lung cancer or pneumothorax, you can. Or, if you want to get down to the nitty-gritty and seek out knee MRIs of patients in the Eastern US aged 30-60 with a torn ACL who had their scans taken on a Siemens machine - you can do that too.

Once a user has built out their desired dataset, they can request a sample for Q/A purposes. We will deliver them a ZIP file with both de-identified reports and DICOM files (which contain the imaging information). If they are happy with this, they can then place an order and pay for their entire dataset, which they can get in hand within days.

We have ~100kLOC of Golang to do this fast (datasets are hundreds of GBs), and a bunch of Python to do ML at scale (classification, de-identification, OCR). We are on AWS. We are PostgreSQL / UNIX fans here, and we're looking for a lot of smart people (data engineers, data scientists, ML experts, hardcore software engineers and systems admins). Usual day is: how to extract million images without any PII out of the hospital and present it to clinical data scientists quickly.

Apply: https://www.segmed.ai/careers

Tech stack: Go, Python, AWS, RDS PostgreSQL, Redis, Airflow, ML

I'm a cofounder/CTO so DM me if you have questions.

Doesn't work. Page not found

How do you NOT pay more for running double of everything + load balancers?

I applied 3-4 times for PM roles at GOOG/FB/etc and each time my resume was auto-routed to folks recruiting for a senior software engineering jobs. I think what you say makes sense, so just don't give up on applying. My $0.02: of course I don't know for sure, but having some end-to-end product that you've made with UI etc. could help.

Mea culpa. It looks even more awesome now. I'm in -- and I think I'll send them a $1 via paper envelope now.