Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
christiansafka
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
5 ms
·
1.
▲
by
christiansafka
16d ago
now do craigslist for human skills, curated by AI!
2.
▲
by
christiansafka
19d ago
Interactive real-time world models. I'm experimenting with different strategies to make them "feel good" to use, as opposed to walking through glue. Also launching pinch.now (main quest) -- which provides instant real-time i
3.
▲
Show HN: Live audio dubbing from any browser tab
(pinch.now)
3 points
by
christiansafka
23d ago
|
0 comments
4.
▲
The Importance of Whimsy
(christiansafka.com)
2 points
by
christiansafka
2mo ago
|
0 comments
5.
▲
Benchmarking real-time voice translation
(startpinch.com)
2 points
by
christiansafka
3mo ago
|
0 comments
6.
▲
by
christiansafka
5mo ago
That is a good looking homemade burger
7.
▲
by
christiansafka
6mo ago
Server-side yes - the app ships with its own open-source remote daemon that gets setup when you connect to the server
8.
▲
by
christiansafka
6mo ago
A native Mac app as a modern tmux replacement. I manage agents and terminal sessions across several GPU machines and having it all in one UI has been great for reducing cognitive load. If anyone's interested in being an early adopter
9.
▲
by
christiansafka
6mo ago
If anyone is looking for a slightly higher level management of persistent sessions, we just launched https://haventerminal.com for Mac
10.
▲
by
christiansafka
6mo ago
I would also describe haventerminal.com as making tmux pretty and usable! but without any setup steps, first class support for agents, and management of multiple machines. Disclaimer: It's a macOS app we just launched and it ships i
11.
▲
Show HN: Haven – persistent SSH sessions without tmux
(haventerminal.com)
4 points
by
christiansafka
6mo ago
|
3 comments
12.
▲
Exploring JEPA for real-time speech translation
(startpinch.com)
48 points
by
christiansafka
7mo ago
|
9 comments
13.
▲
by
christiansafka
7mo ago
We've published our learnings and results as part of our entry to the Interspeech 2026 Audio Encoder challenge. Code and model will be released on the same url
14.
▲
by
christiansafka
1y ago
Found a post from Twilio using the same tech stack in 2013! What a trip. Cool project! https://news.ycombinator.com/item?id=5409914
15.
▲
NextJS-RAG – opinionated SQLite RAG
(github.com)
3 points
by
christiansafka
1y ago
|
1 comments
16.
▲
by
christiansafka
1y ago
Over the weekend I built a website for teachers and parents to ask questions from Maria Montessori but couldn't fit all the relevant docs into context windows. The other RAG libraries were either doing too much (IMO) or involved conne
17.
▲
by
christiansafka
1y ago
Thanks for this feedback! We're rolling out an update soon that will allow you to hear your translated voice without joining a call. CPU usage is on the radar, hope to make some progress there soon as well.
18.
▲
by
christiansafka
1y ago
Yes, we're planning a Windows launch as well! If you need something today, we also have our own video conferencing platform with AI interpretation built in that works in-browser. It's at https://startpinch.com/me
19.
▲
by
christiansafka
1y ago
Our first focus is desktop, but we'd love to get a mobile app out (or let someone else build one with our API)
20.
▲
by
christiansafka
1y ago
Yes, if you're watching TV on your computer the app should translate all the speech. We haven't done any source separation work (musical singing vs actual speech) so if there's music it may pick up some of the lyrics.
21.
▲
by
christiansafka
1y ago
Thanks! We're planning to add support for a lot more languages over the next two weeks.
22.
▲
by
christiansafka
1y ago
Fun idea, are you thinking about a language learning use case?
23.
▲
Show HN: Pinch – macOS voice translation for real-time conversations
(startpinch.com)
65 points
by
christiansafka
1y ago
|
23 comments
24.
▲
by
christiansafka
2y ago
Yes, right now we're ranging from 0.75-3 seconds for the translation to start, and we're hoping to move the average time lower with our next updates. There will always be some limitation to how fast we can translate (different la
25.
▲
by
christiansafka
2y ago
We're hoping companies with international teams will switch over fully (we have internally), but our initial goal is to attract a subset of the market that has cross-lingual needs and unblock them as much as possible from using it more
26.
▲
by
christiansafka
2y ago
Fantastic question. Our opinion on this is that the higher-bandwidth we can make the communication, the more useful it will be. The reason we've moved from IRC->VoIP->Video is because of the efficiency of information transfer a
27.
▲
by
christiansafka
2y ago
Thanks, there are certainly a lot of fun and challenging problems to take on in the space. On scaling, the agent architecture isn't limited to one machine, so you can also autoscale your machines. It's essentially python's
28.
▲
by
christiansafka
2y ago
It's hard, don't do it :D We have a few supply chain companies trying us out though! Would love to hear more about your experience.
29.
▲
by
christiansafka
2y ago
Appreciate the feedback. On the video side, we currently synchronize it to play out with the translated audio (as often as possible), matching when you started speaking to the moment the translated audio starts. Mentioned in another comme
30.
▲
by
christiansafka
2y ago
Yes! LiveKit is great - and we are using livekit agents but had to override a few low-level library components for our use case.
More ›