Ask HN: What Are You Working On? (June 2025)

Still working on https://periplus.app, and recently started to see some traction.

It's an environment for open-ended learning with LLMs. Something like a personalized, generative Wikipedia. Has generated courses, documents, exams and flashcards.

Each document links to more documents, which are all stored in a graph you grow over time.

wow I just tried this, absolutely fantastic. I really hope you take this all the way, I will be sharing with friends!

Edit: upgrading my review from fantastic to probably one the best first experiences I've had with an LLM app. You got my money!

Do you have any socials? Would love to keep up with updates about this project

Thanks for the positive feedback (and the sub)!! Means a lot.

No socials so far as I've mostly been posting updates on the Anthropic discord. But I made an X account for it just now (@periplus_app) where I'll mirror the updates.

You can also reach me any time by email for bug reports, feature reqs etc.

Supremely impressive, and I lean a bit towards the more AI-hesitant side.

This is great. I love this concept. Built something similar myself a few months back (just the course generation part): https://quickguide.site/

A few courses I generated using above:

- https://dev.to/freakynit/network-security-cdn-technologies-a...

- https://dev.to/freakynit/aws-networking-tutorial-38c1

- https://dev.to/freakynit/building-a-minimum-viable-product-m...

- https://dev.to/freakynit/startup-metrics-5ed7

I tried to get it to generate a foreign language reading comprehension course (and even included custom instructions to make the course generate reading comprehension passages to emulate a test), but it just generated a course about _how_ to effectively read different kinds of texts, without actually generating the foreign-language passages themselves.

Yeah, doesn't work for generating language-learning content yet. Something more aligned to what you'd find on Wikipedia tends to work best.

I'm thinking you could have it in the same interface eventually, but right now all the machinery & prompts assume it's decomposable declarative knowledge.