DEV Community

jakeyoung
jakeyoung

Posted on

I’m building Readua — an AI-assisted reading app for deep reading (beta invites)

I’m inviting a small group of people to try Readua, a reading app I’m building to help you read deeply with the help of AI — without replacing the act of reading itself.

At a high level, Readua is “just a reading app”. But my goal is bigger: I want it to cover the reading situations people actually have every day — text, audio, and video; web pages, PDFs, and EPUBs — so when you want to truly read something (not just bookmark it), you’ll reach for Readua.

Since LLMs took off, I’ve been using them to improve both the speed and depth of my understanding. The effect has been surprisingly real for me: it feels like the limits of language and cognition get pushed a little further, and this wave seems more fundamental than the previous ones.

Naturally, I wanted to turn that into a product.

At the same time, I saw a lot of “AI summary” apps and browser extensions that give you sliced, extracted “highlights”. They’re useful, but they don’t replace reading. For many topics, reading the original text is unavoidable — and I still couldn’t find a product that matched how I actually want to read.

So I started building it myself.

Two goals I’m focusing on

  • Make the full reading experience great. If the reading experience isn’t solid, everything else is noise.
  • Explore what “AI-assisted reading” should really feel like. This isn’t something you solve in one shot — it needs iteration with real readers.

Last month I read a post from Andrej Karpathy on X, and it matched my own habit really closely: do a first pass manually, then use the LLM for explain/summarize, then Q&A to close gaps.

https://x.com/karpathy/status/1990577951671509438

I’m starting to get into a habit of reading everything (blogs, articles, book chapters,…) with LLMs. Usually pass 1 is manual, then pass 2 “explain/summarize”, pass 3 Q&A. I usually end up with a better/deeper understanding than if I moved on. Growing to among top use cases.

Build in public

I really want to build something that’s genuinely useful to other people. I’ve talked with a bunch of folks about how they read and where they get stuck, and I’ll keep listening and shipping based on what I hear. Build in Public is something I plan to keep doing consistently.

If this resonates with you, I’d love for you to join the beta list:

I’ll send invites as new features land. Web and desktop versions are coming soon as well.

Top comments (0)