Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
arbayi
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
121.
▲
by
arbayi
1y ago
https://github.com/dashersw/liquid-glass-js maybe you can also check this out?
122.
▲
by
arbayi
1y ago
I’ve even thought about building my own DIY e-ink reader/note-taker for this. How does apps for Paper Pro works?
123.
▲
Ask HN: E-ink devices with real AI/LLM integration?
4 points
by
arbayi
1y ago
|
2 comments
124.
▲
by
arbayi
1y ago
From the original linkedin post: "AI has a product problem. Not a model problem. Models are making capability leaps every few weeks but AI-native product innovation hasn’t kept up. Most products are forcing AI into existing UX patterns
125.
▲
Alterego: The first near-telepathic wearable
(twitter.com)
9 points
by
arbayi
1y ago
|
0 comments
126.
▲
Edge AI Foundation
(edgeaifoundation.org)
1 points
by
arbayi
1y ago
|
1 comments
127.
▲
by
arbayi
1y ago
https://www.edgeaifoundation.org/livestreams
128.
▲
by
arbayi
1y ago
https://store.vufine.com/ I think vufine might be what you are looking for, I don't know much about them as well, also found it here on Hacker News.
129.
▲
MentraOS – open-source Smart glasses OS
(github.com)
206 points
by
arbayi
1y ago
|
122 comments
130.
▲
by
arbayi
1y ago
I just found out about them and it seems super good. I wonder why Meta doesn't support such a thing called "Meta Glasses Application Store"
131.
▲
WordPecker: Open-Source Personalized Duolingo
5 points
by
arbayi
1y ago
|
0 comments
132.
▲
by
arbayi
1y ago
https://github.com/google-ai-edge/gallery A gallery that showcases on-device ML/GenAI use cases and allows people to try and use models locally.
133.
▲
by
arbayi
1y ago
Hello! I've been playing around with agents, MCP servers and embedded systems for a while. I was trying to figure out the best way to connect my real-time devices to agents and use them in multi-agent workflows. At OpenServ, we have an
134.
▲
Agent built in 3 minutes and used in multiagent orchestration
(twitter.com)
4 points
by
arbayi
1y ago
|
1 comments
135.
▲
by
arbayi
1y ago
Repository: https://github.com/enessusan00/openserv-mcp
136.
▲
Show HN: We Asked 10 LLMs – A Daily Game Where You Guess What AI Thinks
(github.com)
1 points
by
arbayi
1y ago
|
0 comments
137.
▲
Show HN: OpenServ – The AI Agent Playground for Builders
(twitter.com)
2 points
by
arbayi
2y ago
|
0 comments
138.
▲
OpenServ: Multi-AI Agent Orchestration Platform
(openserv.ai)
3 points
by
arbayi
2y ago
|
1 comments
139.
▲
by
arbayi
2y ago
I've been experimenting with OpenServ, a platform for managing multiple AI agents working together. It lets you create teams from an agent pool to handle specific tasks, and you can build agents via their UI or TypeScript SDK. Each age
140.
▲
Show HN: TickerPulseBot – Agentic crypto analysis bot for Telegram
(github.com)
1 points
by
arbayi
2y ago
|
0 comments
141.
▲
by
arbayi
2y ago
Here's a list of all the apps (built by them!) mentioned in comments: - LangTurbo (by @sebnun) - langturbo.com : Learn through podcasts with transcriptions and contextual word definitions - Nuenki (by @Alex-Programs) - nuenki.app : Bro
142.
▲
by
arbayi
2y ago
Harika görünüyor! Are you planning to share it open source? Not: Bu benim en sevdiğim şarkılardan biridir!! (it's one my fav songs)
143.
▲
by
arbayi
2y ago
Hi!!! Thank you for being the 100th* star and for suggestion! I've added this to priority list. *thanks again!!
144.
▲
by
arbayi
2y ago
Thank you! The app doesn't have such a feature yet, but I think it will be needed eventually, and it has been recommended before. I will add this to the roadmap.
145.
▲
by
arbayi
2y ago
Would be cool to have such a tool, maybe an extension using chrome's built-in AI APIs?
146.
▲
by
arbayi
2y ago
I wonder if there are any benchmarks specifically designed to evaluate LLMs' performance in language learning tasks
147.
▲
by
arbayi
2y ago
Would agentic workflows come in handy in these cases? I mean having a controller agent after the sentence is created, where this agent would be able to search the web or have access to a database? or personal notes and ensure everything is
148.
▲
by
arbayi
2y ago
Thank you! I think that would be a good feature. I use the app, and it works fine when I add words as I encounter them, but when I later revisit the list, I sometimes wish I could see the sentences where the word were used (originally) to b
149.
▲
by
arbayi
2y ago
Thank you so much to everyone contributing to this thread! I learned a lot here just by sharing this - the power of open source, I guess! From all those conversations and recommendations, I've gathered a list of features that will hope
150.
▲
by
arbayi
2y ago
I actually want to learn German, but I want to learn it by reading German texts and starting from zero, even though that makes it challenging. I need to look up definitions and such, but translating the entire page defeats the core purpose
More ›