Posted by david927 3/30/2025
Ask HN: What are you working on? (March 2025)
I quit my job a couple years back to work on this app full-time, as well as its companion flashcard app, Manabi Flashcards. The goal is to help you learn through immersion.
What's special about it? Manabi Reader became popular as an Japanese-focused alternative to services like LingQ in that it locally tracks and analyzes all the words and kanji you read and study. It shows you which words are new and which you're currently learning via flashcards, so you can easily find content that suits your level and see what flashcards to prioritize adding. It also passively accumulates an on-device (and in your personal iCloud) corpus of example sentences from your reading.
I had built this part-time while working over many years (starting with flashcards and then the reader app) but going full-time gave me the time to do a full rewrite: SwiftUI, native iOS + macOS, and an offline-first architecture that syncs with iCloud and my server in the background.
Although it has a companion SRS algorithm (SM-2) flashcard app, it's also excellent for mining Anki cards. This works with AnkiMobile on iOS and AnkiConnect on desktop.
You can use it like a web browser for the web, or subscribe to RSS feeds. It comes with a bunch of curated content by level. Recently I added EPUB support, pitch accents, and note-taking with todos.
I'm now almost done adding a manga mode via Mokuro.
Next I plan on adding more media types (video, YouTube, PDFs), AI functionality (grammar explanations, document Q&A, etc), Yomitan/Yomichan dictionaries for bilingual/monolingual EPWING and Wiktionary support, and more service integrations such as 2-way sync for WaniKani, JPDB, and existing Anki decks. I've begun work on these items and hope to share more soon.
I'd also like to make this app much more beginner-friendly so that people with zero Japanese knowledge can start learning. Currently it assumes you can read kana at least.
Previous discussion [2023]: https://news.ycombinator.com/item?id=36674259
We're offering online memberships, event management, and a member database packed with features. Membership management is a crowded space, but it's also a low-tech space with lots of sleeping giants not willing to iterate on their product.
It's been a really fun project so far and even more rewarding to see clubs using embolt for their daily operations.
So far I've built stuff like automatically created and advanced tournament brackets with group stage, match schedules, participation queue and achievements. Not that hard, really, but helped me get into the backend side of web-development.
We currently organize 2-4 events each weekend for 300 people of different skill levels.
So far it's exclusive for our small community, but I'm thinking of offering it to similar groups of people in other cities.
vimgolf.ai
Right now, only has two levels but I soon plan to add all the Vim motions and use reasoning models as bots that start off the level with you. Apparently, reasoning models like o3-high, Claude 3.7 thinking, and Gemini 2.5 pro are good at finding new ways to transform files using Vim. Kind of silly to have them do that, but I find it kind of cool.
I've been working on a new software product (native Windows) that is for analysis of SQLite databases. It's geared towards non-technical and slightly technical who may not know anything about SQL. The software includes an ER diagram, ability to browse table data, query building, and charting (bar, column, histogram, line, pie, scatter). Trying to get it finished up so that it can be released (hopefully in next few weeks).
Working on In or Out. Anyone with Elixir Phoenix experience are welcome to help with the project.
Side project for Fantasy football offline drafting. We aren't allowed any electronics just up to date print outs. However drafting takes forever with stickers. I'm making a Elixir Phoenix liveview app to replace the board and player picks with timer and fun prizes.
The goal is to build a discovery system/algo that surfaces stickiness and fun to give developers a tighter publish/iterate feedback loop so they can really hone their craft on a shorter time frame.
If you have a prototype of a game that can be hosted as a static frontend web rotting away in your "projects/" directory feel free to toss it up on the site. Bugs beware :)
High quality translation requires context. Mainstream translation apps (Google Translate) guess based on probabilities based on what the user would commonly mean, but this can results in confusing translations.