building multiplayer agents @heyollieai. co-founded google assistant for kids & families; scaled youtube tv + snap, early nest + vmware. startups, football.

California, USA
Joined July 2008
crazy to think ab how quickly agent land has unfolded all these launches in last ~10 wks > Grok @bot 8/11 > @instinct 8/26 > @Muse 9/8 > Dots by @OpenAI 9/29 > Duet by @AnthropicAI 10/21
3
7
1
21
4,400
Rushabh Doshi retweeted
The REA (Reverse Engineer Anything) / Reverse Engineer Everything movement is cooking
7
7
1
62
4,139
Rushabh Doshi retweeted
Nikita Bier
@nikitabierOct 12
Technical moats are mostly over at this point. If you’re building something that is meant to be durable, the only moats now are: • Proprietary data • Network effects • A physical component
1,150
2,084
902
25,786
3,317,349
Rushabh Doshi retweeted
To all the people visiting SF for tech week.. 1) the best builders are probably at their own offices. so DM them and try to meet them there. 2) don't be a tourist writing.nikunjk.com/p/stop-b…
20
5
6
214
19,811
Rushabh Doshi retweeted
After 9 years. 16K+ family homes. 500M sqft inside real homes covered, we know home robots are being built in wrong order. So we built Matic the way nature grows a child. Finally, sharing our vision: 1.⁠ ⁠Why robots get stuck at the demo 2.⁠ ⁠Why labs can't ship robots 3.⁠ ⁠Why we started with eyes, not hands 4.⁠ ⁠The 99.999% rule in robotics 5.⁠ ⁠Where Matic is going next 1. Why most robots get stuck at demos Robot demos look brilliant in a room arranged around its limitations: familiar lighting, predictable furniture, nothing troublesome on the floor. A family’s home makes no such accommodations. There are cables, toys, shifting rugs and, occasionally, dog poop. Yes, humanoids will need to avoid that too. Repeating a task under familiar conditions of uncluttered floors is not the same as handling an unfamiliar home with all its chaos. 2. Why labs can't ship robots SLAM (Simultaneous Localization and Mapping) is how a robot maps its surroundings and locates itself within them. Ask most academics and they'll tell you it's a solved problem. But real homes are the most chaotic spaces that exist. There are glass doors that look like open doorways, mirrors, stairs, rugs, charging cables, and Legos placed all around. So ask yourself: if indoor mapping and navigation is "solved," where are the robots? Why aren't airports, hotels and grocery stores full of them? Calling it “solved” misses the question families actually care about: can I leave this thing alone and trust it? 3. Why we started with eyes, not hands Nature doesn't give birth to a fully working human. We took our cue from how nature develops humans: capabilities built on one another. Children learn to see before they learn to grab, and to grab before they learn to plan and handle the unexpected. Perception comes first. Manipulation follows. More complex responsibilities come after that. At every stage, the robot must earn the next job by doing its current one well enough. Before asking a robot to pick up a sock, we wanted it to understand where it was, where the sock was, and how to reach it. Most of the industry started at the top, with humanoids. We started at the bottom, with a robot whose job is to see and move through a home precisely with 1cm accuracy in any lighting condition. Floor cleaning gave that foundation an immediately useful job. It forced us to confront navigation, clutter, privacy and everyday reliability before reaching for more complex chores. The floor cleaner isn’t a detour from our larger ambition. It is how we’re building toward it. 4. The 99.999% rule in robotics At 90%, one in ten decisions is wrong, and a robot makes thousands in a single clean. That's the robot that bumps, gets lost, and falls down the stairs. Every extra nine is a new mountain. The failures get rarer, harder to find, and exponentially harder to fix. They barely even happen until they happen in your home. Matic's visual SLAM runs at 99.999% in 16K+ homes. We’re the ONLY unsupervised home robot at scale with pure vision-only full autonomy. If you ask us, the gap between 80% and 99.999% is where our 9 years of engineering went. We close that gap inside real homes. When Matic handles something imperfectly, it saves a short clip and keeps it on the device. It only leaves if the family chooses to share it. If they do, it gets labeled, fed back into the model, and every Matic gets smarter, including theirs. Families control what leaves their homes. We do the work of making the product better. Better robots earn trust > trust brings more homes > and more homes teach the robot more. 5. Where Matic is going next Phase one was perception and it's nearly done. Matic has covered 500 million square feet, 400K miles in thousands of real lived-in homes. Now phase two is going to be about manipulation. And our idea is to build a robot that doesn't just move through your home, but acts in it to eliminate even more chores. Each step must be useful today, not justified by something we promise tomorrow. The evolution of Matic is the revolution. The goal isn’t to put the most impressive humanoid in your living room. It’s giving your family time and energy back through robots that earn your trust, protect your privacy, and help without becoming another responsibility. More time for each other. Less work getting in the way. That’s Matic. Get yours today at maticrobots.com
143
222
100
1,274
726,110
Rushabh Doshi retweeted
rabois famously classified people in a company as “barrels” and “bullets” in an agi regime there is no room on a team for people to be bullets, that is what agents are for everybody on your team has to be a barrel
33
34
6
524
56,804
Rushabh Doshi retweeted
2 things I always dread: 1. Checking in for flights. The pages of checkboxes and confirmations, re-entering all my info for the millionth time, declining the upgrade offers. 2. Replying to LinkedIn spam messages @heyollieai just handled both of these for me in one quick text
4
1
8
1,512
Rushabh Doshi retweeted
Replying to @faruk_parhat
Hard work society > hot take society
1
3
1
26
37,828
Rushabh Doshi retweeted
The key to Specs is how amazing the experience is. It’s easy to critique how something looks on the outside but we all know it is about what’s on the inside And the inside here is magic. Kudos to the Snap team for persistence and focus to make something great
All of the updates from the launch of SPECS. 🧵
19
17
3
163
41,228
Rushabh Doshi retweeted
Nikita Bier
@nikitabierSep 16
I recently got a dog and I quickly learned that a puppy needs to be taken out every hour to pee for up to 6 months. Otherwise, it won’t ever be trained. I found a moisture sensor that sounds an alarm that could train the dog—but it was a tiny strip that didn’t cover much floor space. So I asked AI to find the engineer who built it, it reached out to him, he shared his supplier in China and sent the original order documents, then the bot drafted the larger version along with a CAD model and it’s getting manufactured now. Software problems are solved. Atom [and dog] problems are next.
742
271
97
8,651
691,650
Rushabh Doshi retweeted
Replying to @illscience
Those are great questions to ponder. A few thoughts from what we’ve seen building in this space. - Most consumer purchases aren’t supply-constrained, so I think discovery, comparison, and convenience will drive most of the value from personal agents. Merchants will naturally apply the same optimization playbook they’ve used for humans to agents: better pricing, loyalty rewards, availability, and convenience. - The underlying models are still RL'ed toward a goal, so I’d expect human preferences to matter a lot. Some people optimize for lowest price, others for value, quality, or brands they trust. The interesting question is how faithfully agents learn and represent those preferences over time. - I haven’t seen the “agent in a group chat” pattern really work yet. We’ve tried it, and so have others, including ChatGPT. My current hypothesis is that group chats are inherently multi-purpose and socially fluid, which makes it hard for an agent to know when it should participate versus stay quiet. - I’m increasingly convinced consumers don’t want to manage multiple agents. Delegation itself is still a new behavior for most people, so managing even one agent well is already a learning curve. I see a similar dynamic in Grok bots, you might start by creating 20+ specialized bots, but eventually you want one or two “chief of staff” bots that coordinate the rest. - iMessage is a great gateway to consumer AI agents. The tradeoff is discoverability and product surface area. Consumers still think of products in terms of “downloading an app,” and messaging gives you much less room for education and exposing what the agent can actually do. Give our product @heyollieai a try. @joshelman and @venturetwins have been extremely helpful to us, and we'd love to hear your feedback on our product too.
1
3
9
386
Rushabh Doshi retweeted
Replying to @alexandr_wang
Same with @heyollieai. Dads and grandparents are welcome too.
1
5
106
Rushabh Doshi retweeted
Helpful traits for the future: - Desire to think for yourself - Driven more by intrinsic validation - Tolerance for ambiguity - Seeking wisdom not just intelligence - Viewing relationships w/o hierarchy These traits always mattered. They are about to matter a lot more.
43
57
8
480
25,520
Rushabh Doshi retweeted
A reservation at Carbone is the least of my AI asst needs. My full plate looks like 5+ peoples schedules to coordinate: two young children with different interests, one of whom has specialist appts + endless forms, a babysitter, a dog, 2 working parents, 2 homes and elderly parents. Zero complaints, very fortunate, still a lot for me + many women like me. I capture everything in Stream Ring @sandbar as I'm moving through my day + it goes wherever it needs to from there. s/o @heyollieai for helping me on BTS scheduling coordination this year.
1
5
354
Rushabh Doshi retweeted
The new moats are the same as the old moats Every few years, we fall in love with shiny new tech and forget the basic physics of consumer software. We’re doing it again with AI. The new moats aren't new at all. They're the exact same as the old moats: network effects, marketplaces, and platforms. Right now, consumer AI is booming. New agents like Instinct, Bot, and Tomo are dropping mind-blowing experiences. The underlying tech is incredible, but almost every product being built today shares the exact same challenge: They are completely single-player. Single-player products are 100% tied to value - and in this case mostly agent : model performance. If a competitor drops an agent tomorrow that books travel faster, tracks habits better, or handles life admin more reliably, everyone can switch overnight because leaving is easy and has nearly zero friction. Especially when it is so easy to onboard with just a new message. The legendary consumer tech giants didn't win because their underlying technology stayed marginally better forever. They won because of structural lock-in: Social Networks: You don't abandon WhatsApp for a prettier UI if your friends aren't there. Marketplaces: Airbnb, Doordash, and Uber hold supply and demand in a tight loop. Platforms: Apple and Android deliver you a complete device so you take advantage of the software on top of it (though this creates opportunities too) Novelty gets you initial distribution. Multi-user dynamics give you long-term retention. If your consumer AI product doesn't become exponentially more valuable to User A when User B joins, you don't have a moat, just a temporarily superior feature set. We are seeing this in the coding agents as people jump from tool to tool based on the best performance. But… all is not lost. There are huge opportunities here. Agents will get better when more of our friends are on them and can help us coordinate and communicate to do more together. Agents that help us improve and strengthen our habits can get better as we add friends and hold each other accountable. Data flywheels are great, but social and marketplace flywheels are what actually build enduring tech giants. It’s time to stop building isolated AI tools and start building the platforms where people connect, transact, and coordinate together.
195
163
95
1,773
276,998
Rushabh Doshi retweeted
what are the main products in the personal/professional assistant space? more professional - chatgpt work - claude cowork - claude tag - grok bot - manus (and claude code and codex) more personal - town - tomo - instinct - ollie - pallie drafted this 2x2, right axes?
57
18
7
217
38,566
Rushabh Doshi retweeted
Prediction: Instinct will be free to use.
1
1
1
3
1,706
Rushabh Doshi retweeted
I worked at Snap for 8 years so this would be easy to answer from experience.
SITUATION BREWING: Anthropic is asking prospective employees in culture interviews how they would feel if the stock hypothetically went to zero due to a significant change in course, per Axios.
74
87
11
6,673
901,467
we’re having debates about datacenters and China is publicly developing superhuman robot armies what the actual hell are we doing here guys
china’s AI robot just hit 14.5 m/s
1,514
2,019
237
22,222
2,894,506
Rushabh Doshi retweeted
It's very common for founders not to know how well they're doing. In both directions. It's obviously common for founders to be making mistakes and not realize it. But it's also surprisingly common for them to be doing great and not realize it.
97
81
4
1,991
179,226