kache

@yacineMTB • 330,319 subscribers

reinforcement learning, robots. prev eng @ x, stripe. 6'3 (height) tensorpunk subscribe to read my blog!

Shorts

I wrote my own CAD software whose intended user is an AI. I asked GPT 5.6 sol to make me a case for my camera. It only interrupted me once, which was to log in to a website to download a model of my camera as reference. Not perfect but.. pretty crazy :)

I wrote my own CAD software whose intended user is an AI. I asked GPT 5.6 sol to make me a case for my camera. It only interrupted me once, which was to log in to a website to download a model of my camera as reference. Not perfect but.. pretty crazy :)

92,444 次观看

Dennis Gustafsson me and gpt 5.5 found a bug that didn't make it deterministic and fixed it

Dennis Gustafsson me and gpt 5.5 found a bug that didn't make it deterministic and fixed it

152,633 次观看

This is running in my web browser on my laptop. It's deterministic and running using webgpu Tonight; I have a /goal going where it is going to set up one of my 4090s as the authority server, and then test all the rollback code. Goal is 50k boxes simmed per user, up to 512 users. Massively multiplayer massively rigid body physics online games inside your web browser. Why the hell not, we've got the tokens to spare!

This is running in my web browser on my laptop. It's deterministic and running using webgpu Tonight; I have a /goal going where it is going to set up one of my 4090s as the authority server, and then test all the rollback code. Goal is 50k boxes simmed per user, up to 512 users. Massively multiplayer massively rigid body physics online games inside your web browser. Why the hell not, we've got the tokens to spare!

36,148 次观看

I just trained cartpole in mujoco at 18 million steps per second. This policy learned in **less than 3 seconds** rollout policy batch size was 8192 agents

I just trained cartpole in mujoco at 18 million steps per second. This policy learned in less than 3 seconds rollout policy batch size was 8192 agents

68,441 次观看

"how many experiments have you run" yes

"how many experiments have you run" yes

45,350 次观看

what if you simulated every single particle of your robot would it be parallelizable then?

what if you simulated every single particle of your robot would it be parallelizable then?

31,586 次观看

researcher friend of mine shared this with me; comfree-sim. which is basically existence proof of my hunch: mujoco has terrible performance, and there are probably huge huge gains you can make in training, especially considering domain randomization

researcher friend of mine shared this with me; comfree-sim. which is basically existence proof of my hunch: mujoco has terrible performance, and there are probably huge huge gains you can make in training, especially considering domain randomization

35,284 次观看

I've gotten a mujoco sim RL training loop for a unitree robot at 200k SPS. I'm looking into the physics for friction, contact dynamics. My goal: can I reproduce & beat the mujoco playground RL baselines This is running in my web browser with raylib. Its the baseline

I've gotten a mujoco sim RL training loop for a unitree robot at 200k SPS. I'm looking into the physics for friction, contact dynamics. My goal: can I reproduce & beat the mujoco playground RL baselines This is running in my web browser with raylib. Its the baseline

29,728 次观看

today; the goal is to reduce the time of training for this by 10x. This took 27.8 minutes to train at 200m steps 120k steps per second. reproduction from mujoco playground using mjwarp

today; the goal is to reduce the time of training for this by 10x. This took 27.8 minutes to train at 200m steps 120k steps per second. reproduction from mujoco playground using mjwarp

27,376 次观看

6 pendulums probably in 5 hours

6 pendulums probably in 5 hours

26,064 次观看

I'm building the Canadian DJI with a debit card

I'm building the Canadian DJI with a debit card

33,993 次观看

my first fluid sim

my first fluid sim

35,400 次观看

Fixed a numerics bug on the training loop, and got the training down from 52 minutes to 45 minutes Only 15 minutes away from beating the deepmind mujoco playground baseline!

Fixed a numerics bug on the training loop, and got the training down from 52 minutes to 45 minutes Only 15 minutes away from beating the deepmind mujoco playground baseline!

19,248 次观看

configuring nvim for 5 hours today was worth it look at this cute lil guy

configuring nvim for 5 hours today was worth it look at this cute lil guy

136,084 次观看

i want to show you guys how fast programming is now i'm kind of resurrecting dingboard zig to see how fast i can make progress on it, using it as a test bed to see AI progress it took me 5 minutes to add a GPU accelerated (!) mirror button now, watch how fast this next thing is

i want to show you guys how fast programming is now i'm kind of resurrecting dingboard zig to see how fast i can make progress on it, using it as a test bed to see AI progress it took me 5 minutes to add a GPU accelerated (!) mirror button now, watch how fast this next thing is

138,298 次观看

live on my website again. i'm designing the internet condom. Trying to understand the difference of search and recommendations, figuring out what an MVP can look like, trying to design something that can actually beat all social media networks, something that would destroy them

live on my website again. i'm designing the internet condom. Trying to understand the difference of search and recommendations, figuring out what an MVP can look like, trying to design something that can actually beat all social media networks, something that would destroy them

16,064 次观看

right now my reinforcement learning model is basically doing this and i came up with a solution for it all by myself (by stealing it from pufferlib) that not only solves it, but also helps keep training stable and fast

right now my reinforcement learning model is basically doing this and i came up with a solution for it all by myself (by stealing it from pufferlib) that not only solves it, but also helps keep training stable and fast

60,094 次观看

just added skins. do you guys see this?

just added skins. do you guys see this?

85,568 次观看

well it's getting one of them up..

well it's getting one of them up..

11,734 次观看

Videos

Anya Rossi

sweetdream.ai

SweetDream.ai•Sponsored•Livecam

Watch Anya Live

Anya is streaming live right now! Join her private show and enjoy exclusive content.

Exclusive private shows

1.2k viewers online

Private Show

Join now for exclusive access

Free preview available • Premium content

Fable and I rewrote Box3d entirely in native cuda. It's about 30x faster on my GPU compared to my CPU (depending on the situation). This was my hardest benchmark. I still had to be there, but soon I won't. Cost me about 200 dollars

Fable and I rewrote Box3d entirely in native cuda. It's about 30x faster on my GPU compared to my CPU (depending on the situation). This was my hardest benchmark. I still had to be there, but soon I won't. Cost me about 200 dollars

741,183 次观看 • 14 天前

after having sol rip on it all night, i can do over twice as many boxes without dropping to unusable fps

after having sol rip on it all night, i can do over twice as many boxes without dropping to unusable fps

131,596 次观看 • 10 天前

When I worked at twitter; I got ping'd by someone from the bot detection team asking me if my usage was natural -- I hit one of their new test bot detectors. Then I sent him a video of how I use this website and he was like "oh.."

When I worked at twitter; I got ping'd by someone from the bot detection team asking me if my usage was natural -- I hit one of their new test bot detectors. Then I sent him a video of how I use this website and he was like "oh.."

832,946 次观看 • 1 个月前

behold. THE WORLDS FIRST SIX PENDULUM CARTPOLE SOLVE. Including a sponsor! To solve this task, I built an environment to train an AI. This is what mechanize does, but for larger AIs. Apply! Salaries are up on their page Thank you to mechanize for sponsoring!

behold. THE WORLDS FIRST SIX PENDULUM CARTPOLE SOLVE. Including a sponsor! To solve this task, I built an environment to train an AI. This is what mechanize does, but for larger AIs. Apply! Salaries are up on their page Thank you to mechanize for sponsoring!

494,261 次观看 • 1 个月前

this is GPU accelerated, no circular dependency physics "solver" (there is nothing being solved) each particle gets a cuda thread (not literally but you get what i mean). Then it gets reduced up to what it should resolve to regarding the rigid bodies

this is GPU accelerated, no circular dependency physics "solver" (there is nothing being solved) each particle gets a cuda thread (not literally but you get what i mean). Then it gets reduced up to what it should resolve to regarding the rigid bodies

128,536 次观看 • 1 个月前

everyone is building humanoid robots when they should be building this

everyone is building humanoid robots when they should be building this

1,208,266 次观看 • 9 个月前

scales great! each rectangle here is a set of discs. running on cuda. now to painfully go through all of the weird collision bugs and figure out why they're happening and why it isn't settling

scales great! each rectangle here is a set of discs. running on cuda. now to painfully go through all of the weird collision bugs and figure out why they're happening and why it isn't settling

91,129 次观看 • 1 个月前

my website is now live editable by me. While I edit it, you can see the edits happen live

my website is now live editable by me. While I edit it, you can see the edits happen live

147,188 次观看 • 1 个月前

programming in 2024

programming in 2024

1,652,675 次观看 • 1 年前

I think I'm going to become a game dev

I think I'm going to become a game dev

75,355 次观看 • 1 个月前

2023, 3 pendulums 2025, 4 pendulums 2026:

2023, 3 pendulums 2025, 4 pendulums 2026:

87,281 次观看 • 1 个月前

wrote a quick script to scan through the accounts i've blocked and which country they're posting from

wrote a quick script to scan through the accounts i've blocked and which country they're posting from

376,789 次观看 • 7 个月前

I AM SO SO CLOSE TO SOLVING 6 PENDULUM CARTPOLE just added erection reward and it's almost got it but this task is really really hard. I've been pouring compute at it I'm not leaving my computer seat until I have this solved. it's time to bust out curriculum learning

I AM SO SO CLOSE TO SOLVING 6 PENDULUM CARTPOLE just added erection reward and it's almost got it but this task is really really hard. I've been pouring compute at it I'm not leaving my computer seat until I have this solved. it's time to bust out curriculum learning

60,492 次观看 • 1 个月前

i love this stuff so much

i love this stuff so much

55,933 次观看 • 1 个月前

i wonder when this task becomes impossible to solve

i wonder when this task becomes impossible to solve

47,586 次观看 • 1 个月前

I'm launching dingbotics: speeding up RL for robotics go to dingbotics dot com to check out the latest model I've trained. this will work on your phone's web browser built with mujoco, nvidia warp, pufferlib, raylib & Rich the Kid

I'm launching dingbotics: speeding up RL for robotics go to dingbotics dot com to check out the latest model I've trained. this will work on your phone's web browser built with mujoco, nvidia warp, pufferlib, raylib & Rich the Kid

38,927 次观看 • 1 个月前

your product is 3D printed???????? you didn't even bother to get CARBON FIBRE END CAPS ordered from SEND CUT SEND (3 day delivery)?? for my hardware startup, i've been using SEND CUT SEND, this month's sponsor of my profile. kudos to them for helping support our development!

208,582 次观看 • 9 个月前

dingcad is now open source needed cad software for my startup, so i made it i'll keep it pushed with library and updates. i push straight to master. there is a file called scene.js that will live reload the screen every time you change it. it's openscad without the bloat

dingcad is now open source needed cad software for my startup, so i made it i'll keep it pushed with library and updates. i push straight to master. there is a file called scene.js that will live reload the screen every time you change it. it's openscad without the bloat

157,742 次观看 • 8 个月前

pretty insane that it's been nearly 2 years and no one has made anything nearly as good as this

pretty insane that it's been nearly 2 years and no one has made anything nearly as good as this

180,686 次观看 • 1 年前

my second fluid sim

my second fluid sim

41,239 次观看 • 2 个月前