
building @BodyBuddyCoach - a personal agent for health. ex-eng @ Nerdwallet, Dropbox, Epic Systems
try for free π
β
Joined May 2011
- Tweets3.1K
- Following938
- Followers652
- Likes3.3K
Pinned Tweet
i think we found the missing UI for AI agents in iMessage.
and it's been staring us in the face this whole time...
the challenge is you constantly have to ask what the ai is doing.
so i started experimenting with live updating the Dynamic Island while you chat:
almost every company in the health industrial complex is focused on INFORMATION.
we already have the entire corpus of human knowledge at our finger tips yet many still choose to eat at McDonald's and live sedentary lives.
information is not the bottleneck.
people need help taking ACTION
the 100+ biomarker craze is insane.
for vast majority of people it is simply a roundabout way to tell you what you already know - eat better, move more, sleep consistently.
you are much better off paying a coach and getting daily accountability to actually execute on a plan.
these guys are sitting on this technology that's going to alter the livelihood of so many people and they can't stop vague posting and hyping about it and keep raising people's anxiety about it.
the lack of EQ is insane.
Yeah, I get it. You're excited. Everyone's excited. But a lot of people are also fucking terrified, bro. Like, do you not get that? Like, YOU GUYS are also terrified.
So why would you be hyping it up so mucj?
imagine the Manhattan Project folks being like, "Oppenheimer's fucking locked in bro"
as peter said in the comments let the work talk you don't need to rile the crowd up more
The near future of knowledge work is everyone out and about doing their hobbies and stuff while managing personal agents with their voice
Predicting an imminent rise in sports like golf where you can easily be working all day outside while enjoying life. Would say there at least a few year window where this becomes a standard way of working for many.
The last hurrah of knowledge work
@Supertake Iβd like to long Golf or other leisure sports plz and thank you
I got a concussion surfing this week. Doc says I have to greatly minimize screen time.
So Iβve completely put my laptop away and getting work done with voice via dot or other agent while outside.
Itβs pretty obvious this is the near future of knowledge work and itβs incredible
I dunno how long humans will still need to be in the loop at all but at least as long as we do are getting knowledge work done on a computer while not at a computer is phenomenal.
A good agent will order a pizza for you on DoorDash. A great agent WONT order a pizza for you bc itβs not good for your diet.
We're giving away $1,000,000 of pizza right now all across the U.S.
It's pizza night, on us. To celebrate the launch of Hark Pro. Download the app and tell Hark to order a free pizza. hark.com/download
Instructions in thread to order yours now.
Readers added context they thought people might want to know
Users report that Hark's $1M pizza promotion frequently fails because the $30 DoorDash gift card does not apply at checkout, sometimes resulting in full charges. Hark acknowledged a temporary bug and said affected users would be contacted.
hark.com/pizza
x.com/hark_labs/statβ¦
x.com/adcock_brett/sβ¦
x.com/rickymay/statuβ¦
x.com/brianonorio/stβ¦
Okay Tbf thatβs not why bodybuddy didnβt order anything here itβs more I just havenβt given it the ability but cmon man pizza ordering has been the single most optimized purchase on the entire planet for decades. It would take a neuralink in my brain to order with my thoughts for it to seem like an improvement.
the biggest bottleneck by far is the harnesses. they are so bad. the memory is so bad.
Do any task for more than like 2 weeks with any "agent" and it will totally flop.
they are all a mile wide and an inch deep
"Our initial evaluation suggested that GPT-6-Astra's score was in the late eighties followed closely by Opus 5.5. But, as we started to investigate the mistakes that the models might have made, the models, especially Astra, insisted that they were right and that it was the expert-supplied answers that were erroneous. When we went back to the experts it turned out, in multiple instances, that that the experts found some subtle errors buried deep inside their calculations. As they revised their answers, Astra's score started to inch up and it finally ended up with 98% (49/50)."
I just had Dots do a really disgusting thing of completely disregarding a long conversation I had yesterday with it about how to communicate with other people.
Really really amateur and trust totally lost
healthcare is going to shift very dramatically.
we have a "perfect storm".. technology that is primed to revolutionize combined with an industry known for its inertia, friction, and incremental progress.
most takes i see are not very far removed from current state because thinking far outside the box is just not something you can traditionally pull off in healthcare
We're building the agent Dr. House would want to consult with instead of having to go all detective mode in a patient's house to figure out wtf the obscure thing is in their life causing them the rare disease
(it's lupus. it's always lupus)
Agents are friction reducing machines.
As a coach, we built BodyBuddy to understand internal friction like "why didn't you go to the gym today?"
even a few layers deep into asking why we can identify meaningful frictions in someone's life.
that can look like an injury, low energy, they're busy, poor mindset, lack of planning ahead, etc.
we track the friction and deploy agents in background to do whatever we can possibly do to help them overcome this friction.
one of our key metrics is how well it is actually helping people reduce frictions







