chief code deprecator @virioai prev: ff @southpkcommons, apm @salesforce, eng @stripe, cs @yale
SF
- Tweets 730
- Following 743
- Followers 724
- Likes 4,726
Emmett Chen-Ran retweeted
Epic office warming party last night, thanks to everyone who came through 🙏
Hosting a BBQ at my place this Sunday for founder friends! DM me if you want in
1
2
25
1,256
Been Willowing all afternoon and can confirm: it’s lightning fast
Stop #7 of our voice maxing campaign: @virioai!
We dropped by the Virio office to deliver some mics and got to spend time with the team. They’re using Willow to dictate faster and get more done. Such an awesome group of people, and really cool to see what they’re building (shoutout @doubleemt).
Thanks for having us Virio, really enjoyed getting to know you all!
5
3
12
2,014
Emmett Chen-Ran retweeted
voice-maxxing stop 7: @VirioAI!
visiting the office of the goats of growth hacking 🐐set everyone up with mics that pick up your voice super well when you’re literally whispering
@doubleemt thanks for welcoming us!!
1
2
7
340
Sooooo thankful to not be in the idea maze rn
I understand I’m sitting in the cheap seats as an investor but I really do think Startups as an industry has experienced such model collapse over the past 12-24 months as founders have ceded so many spaces to labs and as the echo chamber has gotten louder, that we are at a local bottom for creativity in new company formation idea space.
It's not that there are no *good enough* ideas, just that they are all quite obvious/imaginable, and thus everyone kind of builds the same thing.
I think that barometer or concept matters a lot.
I like to ask people what the last truly unimaginable or unique idea they heard was and have gotten a lot of blank stares in the past year.
5
664
when me and my cofos were starting out in the idea maze, we thought that "all the easy ideas had already been taken"
which honestly feels preposterous looking back
a company achieving PMF is a unique response to the particular shape of the world at that moment. every second, that shape changes—which means every second, the set of possible companies to build changes
the user behaviors to optimize are different. the backend tech that unlocks new use cases is different.
an "easy" 2004 company might be hard in 2026, and vice versa:
building openai in 2004 might have been easier than building facebook. and maybe in an alternate timeline where AI happened before social media, they're starting to build facebook in 2026!
7
1,222
us: publishes a /thesis
competitor: publishes a /thesis
us: changes our home page hero section to "content that drives revenue"
competitor: changes their home page hero section and LI tag line to "content that drives revenue"
feels good to be no. 1
6
780
realizing how much of technical seniority/experience is just explaining things simply (yet accurately) and concisely
I’ve never been a SWE full time and was brain dead in all of my SWE internships so I never had any really good, really technical mentors
Recently I had a long chat with an exited CTO of an $XXM acq and was struck by how simple and clear a thing he drew on the whiteboard was
progressive disclosure! That’s all it is
Meeting people at their level of abstraction/exposure to the topic and only double clicking into more complexity when asked
I think a lot of technical people feel the need to demonstrate/peacock their technical prowess and try to “out-niche” each other. Plenty of BE eng try to one-up each other in convos by referencing some ever more esoteric thing than their convo partner
But as I’ve interacted with more “staff” eng or CTOs of real cos—technical people who have done excellent technical work—the most common behaviors I see are
1) being fluid and adept at traversing the abstraction ladder of complexity and
2) starting with simplicity by default
It’s a recurring piece of feedback I’ve given all my young engineers on how they communicate: complexity is the easy thing. Simplicity is what’s hard to effect
10
644
tbh wispr's new desktop notetaker is nothing special over granola but is actually perfect for my use case
the main thing I look at granola for is the live transcription in the middle of a meeting, and wispr's transcription is just SO much faster
2
331
Marketers at frontier cos have it so easy. Every launch it’s just “introducing our most advanced ____ ever” hit enter send
1
6
647
Emmett Chen-Ran retweeted
whoever created the opening animation for @VirioAI deff cooked w it fr 👨🍳
2
3
17
1,658
Emmett Chen-Ran retweeted
Today Reactor is coming out of stealth. We’ve raised $59M in Seed and Series A funding, led by @lightspeedvp, with participation from @AmplifyPartners, @wndrco, @Sky9Capital, and @FPVventures.
Reactor is the platform for building in the World Model era: the infrastructure that lets developers build with them at global scale for the first time. Stream from a frontier World Model to your app, in real time, all in under 10 lines of code.
World Models represent the next major shift in AI: pixels, audio and actions are generated on the fly, in real-time, in response to user inputs, and to the environment. Every time computing has made a shift from passive to interactive, entire industries appeared that didn't exist before. We're standing in front of such moment again.
Over the last 6 months, we’ve assembled an all-star team with alumni from Apple, Meta, Google, Luma AI, Netflix, and Replicate. We're already partnering with some of the biggest names and labs in the world, and hundreds of developers are already building on Reactor.
The World Model era starts now.
174
209
2,300
12,738,751
building great product is a lot like being a magician
one has to dance around the limits of expectation and heighten experiences for the viewer such that it feels like magic
4
373
I met Sean at a GC event a few months ago, then my co added $1M ARR in 30 days
Yesterday I interviewed @SeanZCai about AI data.
This is essentially a guide for founders on how to sell data and RL envs to AI labs.
"I've never seen a data contract get turned down by a top lab, if it's good quality data, for budget reasons."
00:00 What areas of data are underserved?
02:10 For bio data, is it real-world or purely digital?
04:21 For cyber data, which subsets are most underserved?
05:50 What is the sales process like?
07:04 Why would a lab not renew or increase their purchase volume?
10:13 When a researcher is exploring a new direction, what's the first step?
11:35 In robotics data, what do you view as underserved?
13:12 What does the initial data delivery look like, what format?
13:53 Do labs have more sophisticated internal setups for running environments?
14:32 Are the non-frontier labs buying off-the-shelf data from Anthropic / OpenAI vendors?
16:11 Do Anthropic data vendors put expiry timeframes on the exclusivity?
16:42 Are purchase decisions researcher-led?
17:41 Decagon, Sierra, Ramp: what kinds of data are they buying?
19:06 Long-term, when do labs still need to buy external data vs train on user traces?
21:15 Will end-vendor benchmarks shift to performance per dollar?
22:04 How many labs are spending at the 1B+/yr data level?
23:53 Delta between Anthropic's stated $1B and your 10-20B/lab number?
26:05 What makes inference providers / neoclouds a good fit to acquire RL env cos?
1
15
7,237