Introducing Second, and a $10M seed led by @a16z.
Second builds a model of you that you own, and your agents can use as an API.
Truly personal software needs a deep understanding of you: not just your emails and messages, but how you work, what motivates you, your people, priorities, style, goals, and intent.
This requires a new kind of personal AI infrastructure. One that continuously aggregates your context, distills it into durable understanding with a ton of compute, and then serves it securely, ubiquitously, and up-to-the-minute. One that builds your *portable user model*.
Second is that infrastructure. It installs seamlessly on your Mac, and your context just comes out. Live APIs of your priorities, to-dos, wikis, writing samples, preferences. Or what you did minutes ago. Or context packs of anything, dynamically assembled in seconds.
Second is streaming-state from local to a private cloud, which means anything you connect to--whether Claude, Dot, Muse, Grok Bot, Instinct, or iMessage--shares the same, growing, continuous understanding of you.
Your portable user model is as powerful as it is sensitive. We founded @FirstPersonLabs on the principle that it should belong to you. This commitment means taking on hard systems and security problems and making sovereignty an invariant of Second from the beginning.
We've barely scratched the surface of what portable user models can unlock. Deep understanding of us, served at machine speeds, could drive the next leap in personal software, human-computer interaction, large-scale coordination of humans and agents, and more!
We're tremendously grateful to be supported by @a16z (@martin_casado, @YODELAGARZA), Mozilla Ventures, @JeffDean, Ram Shriram, Jeetu Patel (@jpatel41), @percyliang, J. Alex Halderman (@jhalderm), @sanmikoyejo, and many others.
Our vision is for a future where every person is more powerful, because they have the ultimate affordance: personal computers that fully understands them on their own terms.
If this excites you, we're hiring and come build with us!
before Starbucks, someone bet coffee could be an experience, not just a drink.
before Disney, someone bet a 30-second cartoon could become a full-length movie.
adaptability and change isn’t just new to tech, but for the economy of tomorrow as a whole.
check out @cirfinpodcast’s latest w/@pejmannozad@pearvc - huge thank you for coming on the pod!
https://t.co/dhi8Z7c8tr
1/8
The real-world is messy. 🤖🧹
If there's no collision-free path, a robot shouldn't just stop, it should move objects out of the way. But which object can it safely move?
Introducing MessyNav: Zero-Shot Navigation via Manipulation✨ Accepted at #CoRL2026 🚀
🧵⬇️
Excited to be at #CoRL2026 next week to present CHORUS!
happy to chat about collaborative, cross-embodiment learning, & feel free to drop by our poster on Monday 🦾
Teleoperators collecting data to fine-tune VLAs are spending valuable time demonstrating actions robots already know how to do.
TANDEM combines task and motion planning (TAMP) with human teleoperation only as needed, reducing human effort.
(1/8)
Is being technical still worth it when AI commoditizes math and coding?
We asked @Alfred_Lin of @Sequoia. We also talked about feature vs. real company, circular money in AI, and the rate of diffusion of technology.
Full talk 👇
https://t.co/0zRdz0zjEh
Want to know how well agents can really control robots directly?
In RoboARCade, we systematically evaluate and diagnose frontier agents controlling robots directly using tens of thousands of dollars’ worth of agent trajectories. RoboARCade contains diverse, precise, and long-horizon tasks. Paper, code, assets and every trace are out today. https://t.co/WCCWpc4V4C
Human fingers move in coordinated patterns. Can those patterns help robots explore?
Excited to share EigenDEXplore! 🤖✋
Key idea: add exploration noise along directions learned from human hand motion, while leaving robots free to discover motions of their own
See @harshgupta's 🧵
huge congrats to @AnjneyMidha! making compute more accessible was such an important part to @frontiersystems, and now to see it happen with such scale is incredible!
hundreds of MI355x and B300 nodes
now live on https://t.co/t0uE0bekz3
subsidized for all .edu and .govs
it’s time to scale the frontier ecosystem
for the public benefit
Task and motion planning (TAMP) excels at multi-step, semantic, and geometric reasoning. VLAs excel at deformable object manipulation, which TAMP can't do. Can one VLA do both?
DATAFARM fine-tunes VLAs on TAMP demos aligned to the pre-training distribution.
(1/9)
🤔 Can we train one VLA policy to control multi-robot teams without any explicit communication?
✨ Introducing CHORUS: a single policy for decentralized, multi-embodiment collaboration
🧵⬇️
My take: It’s true that every form of VLM finetuning suffers from some form of catastrophic forgetting. And so training a model from scratch isn’t 𝘦𝘯𝘵𝘪𝘳𝘦𝘭𝘺 as crazy as it seems, and promising even. But it does feel like making a good use of internet-scale priors can only help, not hurt us — 𝗲𝘃𝗲𝗻 𝗮𝘀 𝗿𝗼𝗯𝗼𝘁𝗶𝗰𝘀 𝗱𝗮𝘁𝗮 𝘀𝗰𝗮𝗹𝗲𝘀.
In particular, one of the biggest benefits of fine-tuning VLMs for robot actions is vision-language grounding.
What does it mean to move the leftmost plate vs the one on the far side of the table? Fold the magenta shirt instead of the turquoise one?
500k hours is a lot, but considering the number of tasks, demos needed per task, and prompt diversity across all of these …. 𝙩𝙝𝙖𝙩’𝙨 𝙖 𝙡𝙤𝙩 𝙤𝙛 𝙙𝙖𝙩𝙖 𝙣𝙚𝙚𝙙𝙚𝙙 𝙩𝙤 𝙧𝙚𝙬𝙧𝙞𝙩𝙚 𝙠𝙣𝙤𝙬𝙡𝙚𝙙𝙜𝙚 𝙩𝙝𝙖𝙩’𝙨 𝙖𝙡𝙧𝙚𝙖𝙙𝙮 𝙘𝙤𝙣𝙩𝙖𝙞𝙣𝙚𝙙 𝙬𝙞𝙩𝙝𝙞𝙣 𝙑𝙇𝙈𝙨.
That said, it is exciting to see new approaches rethinking how we train models w/ large scale robotics data. Kudos to the @GeneralistAI team for making it happen (and demo'ing some pretty cool robot-repair tasks!)