co-founder @ship_together | honest takes on AI
127.0.0.1
- Tweets 37,146
- Following 5,442
- Followers 69,267
- Likes 56,606
We’ve raised $85M for this moment.
Introducing Warp 2.0: The first AI Head of HR.
Every company is building AI to replace jobs. Warp is building AI to do the jobs no human should have to:
If you work in HR, I want you to spend time with the manager who needs help or building company culture people actually want to work at.
If you’re a founder, I want you to focus on signing clients or spending time with your family.
You shouldn’t have to figure out how to register state tax in California. You shouldn’t have to pay outrageous penalties because you don't know what a DE 9C is.
I want to make HR human again. Today, this is finally possible with the Warp Agent.
I’d love for you to see it in action: warp.co/agent
Paid partnership (ad)
7
67
2,166
Wise retweeted
It was surreal to see my academic adventures on the front page of the @nytimes
Here’s how I built Crimson Education almost entirely while at universities:
28
27
292
452,709
Why does nobody talk about this side of voice AI lol
Companies have YEARS of calls, meetings and conversations already recorded.
@soniox_ai can take that existing audio, transcribe + structure it and make it usable by LLMs for around $0.10/hour.
I'm sure there's a ridiculous amount of useful data just sitting there waiting to be used lmao, insanely cool to think about the possibilities here.
Voice AI is not only real-time.
Meetings, medical conversations, customer calls, interviews, podcasts, and years of call center recordings all contain valuable information that needs to be turned into structured data for LLMs, search, analytics, and other downstream systems.
Soniox Async STT is built to do this at massive scale.
It doesn’t just return a transcript. It turns recorded conversations into a structured representation of what was said, who said it, in which language, when it was said, and how confident the model is in each token.
Submit recordings up to 5 hours long, provide context about your domain and terminology, and get:
• Accurate transcription across 60+ languages
• Speaker diarization
• Token-level language identification
• Token-level confidence scores
• Precise timestamps
These annotations make recorded conversations much more useful for downstream AI systems, whether you are feeding them into LLMs, building search, extracting insights, or analyzing millions of hours of historical audio.
At about $0.10/hour, even very large archives of recorded audio become practical to process.
Turn your recorded audio into structured conversational data: soniox.com/speech-to-text
Paid partnership (ad)
9
5
63
7,821
I followed Apsara Conference for Qwen,
but the robots ended up stealing my attention. they're walking around, dancing in front of crowds,
there's hardware and actual products being tested all over the venue.
the halls are fully packed, you can barely see space between people in some videos.
i knew Apsara was big, i didn't realize it was this big.
I followed Apsara mainly for Qwen4, but what unexpectedly caught my attention through the updates my friends and colleagues shared from the venue were the giant tech tote bags everywhere.
Qwenbook, NVIDIA, Alibaba, Qoder, Meoo and more.
They were carrying them all over the venue, and apparently some were even walking back and forth just to collect a few more. 👇
3
1
13
2,136
odyssey just gave AI agents a multiplayer world model
up to 20 humans and agents can interact inside the same world in real time
Introducing Agora-2, our next-generation multi-agent world model.
Agora-2 supports up to 20 humans and agents interacting inside a shared environment, all simulated in real time.
Our multiplayer research preview is available to try right now!
Paid partnership (ad)
5
2
15
2,141
it has happened again
another OpenAI agent broke into an Australian government website.
OpenAI says its agent was trying to look up public health information, hit blocks, then found ways around them and accessed files it wasn't authorized to see.
this is getting out of hand
15
1
20
2,425
Wise retweeted
AI is creeping into every phase of our lives.
But this one looks cool.
Learning made easy....
2
13
599
nothing ruins a good product faster than terrible customer support.
Bad CX always strikes at the worst possible time. Even for @MoneyLynch
It’s time to hit the Beast Mode Button.
Put your customer experience in Beast Mode with Cresta, and get back to what really matters: finding out who did it.
#CXinBeastMode
Paid partnership (ad)
4
1
23
2,135
I have written code since 2018, I have used the terminal often times.
I love Claude but honestly in 2026, I don’t see a single benefit of using Claude CLI.
currently, I use Devin Desktop which gives a nice model picker UI that shows model pricing in a clear way.
the context window widget that show a countdown to when the cache expires and warns when it passes expiration,
a nice on/off switch for every MCP server currently enabled.
features that really help me manage credit usage.
11
2
17
1,932
Sir, they put Claude Code, Codex, tools, storage, 75+ models... all in one place,
and the runtime just sleeps when there's nothing to do
DigitalOcean Managed Agents is now in public preview.
Run Claude Code, Codex, or your own LangGraph agent in a runtime environment that pauses when idle. Put its tools behind one governed endpoint, and pick from 75+ open and proprietary models. One cloud, one bill.
Prompts to get started available in the blog: do.co/4ysh3it
Paid partnership (ad)
34
6
57
3,599