Wow this sucks
Nvda board pushed its founder Jensen and CEO out.
Big mistake
NVIDIA announces CEO transition. Read the release: nvidianews.nvidia.com/news/n…
59
custom made remote control for my computer:
esp32, mic, speaker, scroll wheel and buttons. still need an external power so not fully "wireless"
scroll wheel for switching windows (alt + tab)
one button works as pressing Enter
one button triggers my speech to text tool
one unused button
wiring is still a mess inside and can't close the box. AI is re-designing the case
1
88
Claude made this 3d model for an esp32 enclosure. I releazed its wrong orientation after printing it, but damn 😳
I told AI I need an enclosure, gave it a list of components im using. Apparently it knows the size of these components so it just went off and designed this 3d models.
3
1
129
made a esp32 device to remote control my computer, its got:
- mic for voice input
- bluetooth speaker
- tactile buttons for input (switch window, enter, activate voice control)
2
1
212
grok @bot is my stock broker now. this week it made me $160
month to date it made $600
another $100 is already in and expect to pocket in the next week
i might be able to push to $1000 for the Oct month
my strategy is simple:
1. sell put on stocks i like
2. sell calls on stocks i own
3. watch out for earning dates and play earnings
4. pick a delta i'm happy with
5. sell weekly or monthly
6. if put assigns, buy stock, if call assigns, sell stock
7. rinse and repeat
because this is so repetitive i started using grok @bot to help me scan and manage trades
hopefully it can be fully autonomous one day and only asks me when tough decisions are needed
117
i asked grok @bot to build an option scanner with @IBKR . this can be quite handy
- compare across multiple tickers and expiries
- filter on delta
- rank by annualized return
- you can click on the grid and it'll show the strike vs 1sd move range
- next up will add GEX and put/call walls to the visual
67
this video did 4 million views, and the channel only started 3 weeks ago, already 52.7k subscribers. it's estimated they make $ 1.4k-22k per month.
i've created a free tool to let you make the same kind of viral videos in 30 mins and probably just cost $0.20
it uses claude opus5.5 model to write code generated videos. Yes everything is code, no image, no video generation that's why it's so affordable.
how it works 👇
1. the AI writes code that draws core elements
2. use those core elements as references, draw a scene on canvas
3. draw frame by frame, at 24 fps, a 3min video means we need to draw 4320 images
4. use ffmpeg to put those images in sequence, then you have an animation
5. AI makes music and sound effects
6. AI adds voiceover
7. because AI controls the visuals frame by frame, it can match the music beats and sound effects perfectly with video.
check the link below ⬇️
2
135
Deepseek v4.1 flash 1017 (left) vs opus5.5 (right)
For me opus5.5 is the clear winner. look at the right one, there are more details on the scene and more moving elements, more vibrant color and character moves more naturally.
cost: deepseek about $0.2, opus spent 2% of weekly quota on a $20 plan
177
Grokbot is teaching me option GEX
- it made a GEX dashboard
- taught me how it works
this is its prediction for today, lets see how it plays out:
The $365–$370 strikes carry heavy call and put gamma, so that band is the most likely spot for TSLA to get pinned today. $370 and then $375–$380 act like ceilings.
85
a user sent me this msg yesterday❤️glad people are liking Typewave!
if you play world of warcraft on a controller, try typewaveai.com it turns your spoken words into text and put into game chat - no keyboard/typing needed!
66
Here's the uncensored version, runs locally
No need to thank me
huggingface.co/AtomicChat/Qw…
🚀 Meet Qwen-Image-2.1-Turbo — create and edit images in just 8 denoising steps! Open weights now available!
Built on Qwen-Image-2.1, Turbo is an accelerated checkpoint on the same 7B visual generation architecture.
Fewer steps does not mean lower quality: it still generates strong 2K images from text, and supports continued creation through natural-language edits, from adding accessories to changing a scene.
Start directly with Diffusers: load QwenImage21Pipeline and the checkpoint’s recommended 8-step sampling schedule is ready to go.
🌐 Prefer an API? Qwen-Image-2.1 Pro and Turbo APIs are now officially live. Choose a hosted API for your application, or use the Turbo weights in your own workflow.
🤖 ModelScope:
modelscope.cn/models/Qwen/Qw…
🤗 Hugging Face:
huggingface.co/Qwen/Qwen-Ima…
🔗 Pro API:
modelstudio.console.alibabac…
🔗 Turbo API:
modelstudio.console.alibabac…
What will you create with 8 steps? Share your results with us!
3
3,579
a year ago i built a pipeline for stickfigure video gen
today this vid was generated all by code, software indeed cooked
98



















