Excited to share that @MiravoiceAI was just featured on the front page of @Crunchbase News! 🎉
https://t.co/DnJzKa8gjU
If you've ever tried to collect high-quality data at scale, you know it’s a broken, miserable process. Let’s talk about it. 🧵(1/10)
@bcrussett got the inside deets on Bryan and Alex’s work last month - it's what I had to build myself for organizing agents, but 10x better designed/more thoughtful
It was an example like this back in December that made me realize we need a better permissions model for Claude, one that extends to bash calls and not the just the polite illusion of what it’s asking you or can do. That became Clash, our sandbox and permissions plugin for agents
We are seven months away from the most consequential midterm election in the history of the United States. Meanwhile, we are fighting a war. These are the structural conditions for a coup attempt in which a president tries to nullify elections and take permanent power as a dictator. If we see this, we can stop it, overcome the movement that brought us to this point, and make a turn towards something better.
https://t.co/3EFUACrS3S
Another interesting startups whose fundraise I covered this week: @MiravoiceAI a startup using AI voice agents to conduct long-form phone surveys, raised $6.3M in a seed funding round.
Miravoice has developed an AI interviewer that it says can conduct phone surveys and voice interviews for “precision data collection” without human interviewers. The surveys are long-form and quantitative, with some including more than 120 questions and lasting over 40 minutes. They span open-ended responses, numerical inputs, multiple choice questions, Likert scales and matrix questions.
“Imagine talking to 100,000 people and instantly capturing what they know,” said CEO and co-founder @najain. “We make that as simple as creating a Google Form.”
@Unusual_VC led the financing, which included participation from Neo, @25m_official and angel investors from companies such as Ramp, PubMatic, Atlassian and Google.
if you build a lot of things very quickly, you take on a lot of tech debt up front. if the debt starts getting called in urgently (new bugs, new feature requests), you risk not being able to pay it in time (lose customers).
"I’ve been maintaining Open Source projects for close to two decades now. The last startup I worked on, I spent 10 years at. That’s not because I’m particularly disciplined or virtuous. It’s because I or someone else, planted something, and then I kept showing up, and eventually the thing had roots that went deeper than my enthusiasm on any given day. That’s what time does! It turns some idea or plan into a commitment and a commitment into something that can shelter and grow other people."
https://t.co/NlAzvxsxNP
the correlation between forbes 30u30 and fraudsters makes complete sense
credibility hides suspicion of fraud
and f30u30 yields highest credibility with the lowest effort. pay to apply, get some recs and you’re covered.
I have a private db of Japanese print sales and I used @claudeai to make a report analyzing this month's auctions: https://t.co/uwMCtbjQ17 identifying under-valued lots, rare prints, comparing prints across auctions. Normally VERY hard to get this guidance.
It's been wild to see Olmo Hybrid go from idea to reality in the last few months! Check out model + tech report (lots of experiments / theory). Thanks to a truly cracked team of collaborators across pre/post-training, eng, evals, theory, kernels, etc. for bringing this to life!
Introducing Olmo Hybrid, a 7B fully open model combining transformer and linear RNN layers. It decisively outperforms Olmo 3 7B across evals, w/ new theory & scaling experiments explaining why. 🧵