celltalk

Yes, DuoBook is a language learning app, so I added RSVP as an add-on feature for re-reading the stories in another language. It’s not the main experience.

My idea was that the learners can do a slow turn with the main reader, and then a quick read afterwards if they want to.

Honestly, I spent at least a week for the pivot alignment but at the end I accepted once in hundred words it’s shifted a bit. I might go back into it at some point though.

Very cool! I like RSVP, I actually just implemented it in DuoBook as well.

https://x.com/eonurkara/status/2079867443338449135?s=20

I got the access after redownloading the codex app.

I guess Plus accounts don't get access to Sol? Or is it because I am in Europe?

Super!

Maybe a section on RNA degredation and DNA stability and how it would affect sequencing would be nice.

Also, down stream analyses are largely missing e.g. differential analysis, pathway enrichment. Not to mention newer single cell techniques and their up/down sides. But good start!

As far as I understand the underlying model is not multimodal. Maybe a quite naive question but can't we improve the performance by a joint embedding of EEG and MEG data? If it scales with log-linear data, maybe it would also improve with other data as well.

I basically did the same thing almost one and half years ago and not many people cared, but I still believe that this is the future for computational biology.

https://celvox.co/solutions/axon

Two weeks ago, I got 100/100 of a test from a big company for a first screening without using AI. I was pretty confident that I would pass the first round, even hinted few of my friends, but ended up being rejected with an automated mail… The job market is insane at this point and I am not sure what the recruiters are actually looking for. If the candidate uses AI they’re eliminated, if not they’re eliminated. I guess this is one of these times we read on history books: great unemployement.

me too… this felt awkward: https://duobook.co/explore-stories

Yes, it’s for agents traversing the web universe like photons.

Get ready for Apple introducing the first Macbook with a headphone chip.

Basically, https://duobook.co :)

We’re literally seeing digital evolution in real-time. These are basically primitive life forms such as bacteria evolving just with tiniest differences.

Right now we’re engineering every bit of it to make it better but in the long run this is unsustainable. It’s going to be so complex that even these digital life forms won’t be able to understand their own digital DNAs, like us.

We know we have DNA, we can measure every letter but it doesn’t mean we understand what’s going on our 14 trillion cells and how each and every one of them is regulated.

I think this analogy not only explains us, or digital beings we see today. It explains everything, quite literally. Still it would be amazing to think about these systems from the perspective of biology, and try to understand the parts analogous to existing frame that we already have. Then we might figure out what to optimize better. For instance if we figure out a certain part of a layer corresponds to “genes” then we might find out there is alternative splicing within it. Wild but worth a shot.

I have been working on quite a bit of things… but lately I went back to my hobby project of the last year:

https://DuoBook.co

It’s like netflix for language, where users can select/create their personal bilangual stories.

I had quite a lot of feedback from HN, friends, random people on the internet and trying to solve the common pain points and find my way around to make it geniunely useful.

- Most people said it’s hard to come up with a story, so I added url grounding. Also added buttons (including HN :)) so people can just click click and get their stories at their level with their interests.

- Made sure people can generate stories without ever signing up

- Each word is highlighted while being read, and the meanings can be checked with a tap. I also added an option for users to read the sentence for being checked how good their pronounciation is.

- Benchmarked 7 different models to get the fastest & highest quality story generation (it’s gemini now) and it’s insanely fast. I might share more about it on the webpage because I am an engineer and I enjoy this stuff lol.

- Added CSV import in Use my words so Anki users can just import their words to study.

- Also people can download their stories as pdf so they can send it to their kindles.

- I am working on a ChatGPT app, so people can just say “@DuoBook give me a Dutch/English story on latest Iranian events” within ChatGPT, but I am a bit afraid that it might be costly lol.

Wouldn’t it be ironic if US used open source Chinese models for domestic mass surveillance and autonomously killing people without human oversight… democracy at its best.

DuoBook :)

It’s a great read but is this really the history of oral peptides?

I have build a site for recapping your OpenAI chat history, never leaving your local device.

It's basically like Spotify Wrapped but for ChatGPT (for now).

They should have had Cloudflare on it.

I think there is a big business opportunity here. Make a site that let companies put their status update on local vps for $100.

I am glad my personal site is not affected, what would I do without all those incoming traffic.

This is wrong, a lot of diagnostic labs are actually going for nanopore sequencing since its prep is overall cheaper compared to alternatives. Also the sensitivity for related regions are usually matching qPCR, and it can give you more information such as methylation on top of that.

A recent paper on classifying acute leukemia via nanopore: https://www.nature.com/articles/s41588-025-02321-z/figures/8

The timelines are exaggarated but still it works and that’s what matters in diagnostics.

I built a yearly recap of your ChatGPT usage — like Spotify Wrapped, but for your chats.

Runs on your browser, and open-source, so you don’t have to worry about your privacy.

What do you guys think about DuoBook.co?

I have made few games myself back in the days, still some people play. Most notably https://numbword.com

With this logic, you should also be able to trim the parts that doesn’t have words. Just add a cut-off for db, and trim the video before transcription.

Possibly another 10-20% gain?

Last week, we had a dinner with a big group and as usual the restaurant said we cannot split the bill. So, nobody wanted to pay it, not because of the money but all the hustle of sending tikkies (money requests) afterwards.

As usual... I got annoyed by it and coded a solution for it with AI :)

billsplit.co

Just upload the bill, review the items, and send the link to friends. Then, they can select and pay their parts. It’s free and no registration required.

Super nice, thank you for this post! Based on this I've updated the vocabulary learning scheduling in DuoBook (https://duobook.co) to FSRS based. We should be up to date with the science based learning :)

Can you reach out to us through the contact form? Maybe we can setup a 10-min call, because we’ve implemented a pro tier.

You can change the settings for showing the translation and auto-play.