Pinned Tweet
born in right generation
huge ass neural nets
cuda errors
claude code
iced americano
huge ass neural nets
cuda errors
claude code
iced americano
2
8
212
62,192
highly recommend working w aryan he is one of the most grounded and brilliant person i have ever met.
godspeed!
godspeed!
[HIRING] The next decade is a chance to work on some of the hardest problems that once solved, create exponential economic value in society.
Our maximalist version of this is bringing autonomy to all forms of work bottlenecked by physical labour. We’re looking for high polymathic engineers who want to lead this work as part of our founding team.
We’re hiring for two roles:
- Founding Engineer, Robotic Systems
- Founding Engineer, Robot Learning
Location: HSR, Bengaluru, Karnataka (5 days, WFO)
Starting date: Immediate
Compensation: ₹25–65 LPA (cash + equity)
Experience: 0–6 years (We also take interns if they’re cracked enough)
More details in the form, apply here: binary.so/nJCcTPO
3
0
37
1,660
atharva ☆ retweeted
Mobile dev just moved to the cloud.
Capy can now build and test your iOS and Android apps on cloud phones, powered by @tryrevyl
Capy can now build and test your iOS and Android apps on cloud phones, powered by @tryrevyl
43
25
289
26,110
atharva ☆ retweeted
Robots are getting better at completing tasks. But what happens when they make a mistake?
We fine-tuned π₀.₅ on two real-world manipulation tasks to showcase whether recovery demonstrations can improve policy reliability more effectively than additional successful demonstrations. 🧵
We fine-tuned π₀.₅ on two real-world manipulation tasks to showcase whether recovery demonstrations can improve policy reliability more effectively than additional successful demonstrations. 🧵
64
50
201
22,009
who is going to save research by making a more current version of distill.pub
1
0
23
1,086
RT @jaiselsingh: A common objection to event cameras: "what happens when nothing moves?"
But you can move the camera. Every edge fires, st…
But you can move the camera. Every edge fires, st…
0
1
0
31
atharva ☆ retweeted
meet McQueen.
an autonomous RC car built to test how pretrained driving models from @comma_ai perform in the real world, using a single RGB camera and remote inference.
an autonomous RC car built to test how pretrained driving models from @comma_ai perform in the real world, using a single RGB camera and remote inference.
9
6
61
4,111
atharva ☆ retweeted
As our customers scale to training robot models using millions of hours of our recordings, we needed to build a new kind of data infrastructure.
Terabytes of data delivered in minutes, and the capacity to deliver over a Petabyte per day to each of our customers, to multiple ime. That’s more than 10 Gigabytes a second, delivered into any cloud, wherever our customers need it.
Using our customer portal, customers select precisely the data they want and it starts arriving in minutes at their destination, including AWS, GCP, Cloudflare, Hugging Face and any cloud provider who can handle the ingest speed.
We’ve written about how we worked with @huggingface and @Cloudflare to build this secure, scalable infrastructure. Link in comments.
Terabytes of data delivered in minutes, and the capacity to deliver over a Petabyte per day to each of our customers, to multiple ime. That’s more than 10 Gigabytes a second, delivered into any cloud, wherever our customers need it.
Using our customer portal, customers select precisely the data they want and it starts arriving in minutes at their destination, including AWS, GCP, Cloudflare, Hugging Face and any cloud provider who can handle the ingest speed.
We’ve written about how we worked with @huggingface and @Cloudflare to build this secure, scalable infrastructure. Link in comments.
11
18
136
39,435
atharva ☆ retweeted
i taught a vision-language model to drive trackmania (AO1)
qwen2.5-vl-3b watches the game , reads the frames, reads the car's telemetry and presses the keys. no reinforcement learning, it learns how to play based on 300 expert replays
finishes about half its laps, best lap 24.6s
qwen2.5-vl-3b watches the game , reads the frames, reads the car's telemetry and presses the keys. no reinforcement learning, it learns how to play based on 300 expert replays
finishes about half its laps, best lap 24.6s
3
1
21
502
@GeneralistAI we envision minimalism and modularity go hand in hand hence hacky umi collection devices will lead to faster development loops.
img credit: kiwi
img credit: kiwi
2
0
10
326