1. X
  2. dsa
Log inSign up
dsa
8,734 posts
user avatar
dsa
@dsa
builder @livekit, @x, @23andme, yc07 investor @sundayrobotics, @coreautoai
future
Joined August 2007
247
Following
10.3K
Followers
RepliesRepliesMediaMedia
  • Pinned
    user avatar
    dsa
    @dsa
    Jan 22
    Today is a day I’ll never forget. I grew up in Cupertino. My dad was a tech founder in the 80s/90s. I was in YC S07. LiveKit is my 5th company. The first 4 didn’t work out. I’ve had a lot of advantages — it still took 20 years to get here. Founders: keep taking shots.
    user avatar
    LiveKit
    @livekit
    Jan 22
    We learn to speak before we learn to read. Voice is the most natural interface we have. We just raised a $100M to make building voice AI as easy as a web app.
    Image
    00:00
  • user avatar
    dsa
    @dsa
    Aug 5
    Waymo pick up and drop off is so bad.
  • user avatar
    dsa
    @dsa
    Jul 26
    Mom: what do you think of Claude Tag? Me: haven't gotten a chance to use it. I just heard @jack has a new thing in this direction, actually. Mom: yeah Buzz, i know. 💀
  • user avatar
    dsa
    @dsa
    Jul 16
    🤯
    user avatar
    Tony Zhao
    Sunday
    @tonyzzhao
    Jul 16
    Replying to @tonyzzhao
    We found a general recipe for Solves: scale pretraining, then hill-climb with minimal in-house data. For the first time, one fine-tuning example can teach a new behavior that generalizes. Below: 4 folding strategies, learned from one example each and tested on held-out setups.
    Image
    00:00
  • user avatar
    dsa
    @dsa
    Jul 3
    LLMs that are great for coding agents aren't great for voice agents. So today we're launching our first hosted model: Gemma 4, optimized for time-to-first-token without sacrificing task completion rates. Fable/Opus/GPT 5.X optimize for intelligence over speed. But a voice agent
    user avatar
    LiveKit
    @livekit
    Jul 3
    Frontier models keep getting smarter…and slower. This is bad for voice agents, where every millisecond before the first word counts. So we optimized for speed. Introducing Gemma 4 31B on LiveKit Inference: 🗣️ 381ms to first sentence — 2x+ faster than the next model ✅ 88% task
    Image

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Advertisement
Advertisement