Chawanan? retweeted
This guy built a two-sided trading bot on Polymarket and made $233,695
His algorithm trades short-term crypto Up/Down markets using asynchronous complete-set accumulation and dynamic inventory management:
1. Starts building inventory almost as soon as the market opens
Its first purchase occurs a median of 33 seconds after the market opens
It keeps buying through most of the trading window. In 5 minute markets, its last purchase occurs a median of 223 seconds after opening
2. Adds the opposite outcome to its position
It bought both Up and Down in 88.76% of markets
It starts buying the opposite side a median of 35 seconds after its first purchase. It then keeps switching between outcomes as the market reprices
3. Turns most of its inventory into complete sets
89.82% of the shares it buys can be matched into Up + Down pairs
The average combined cost of a pair is 96.35c, giving it an edge of 3.65c
But this is not pure arbitrage
Only 58.66% of two-sided markets ended up with an average combined Up + Down cost below $1. The remaining 10.18% of shares are unpaired, leaving a directional residual
Its Polymarket account: norm1e69
You can build your own trading bot here:
join.horizon.trade/retrovali…
(A trial is available after signing up)
The algorithm’s main edge comes from asynchronous pairing: it keeps switching between Up and Down, matches most of its inventory into complete sets, and retains directional exposure through the unpaired remainder
Chawanan? retweeted
INSTEAD OF WATCHING NETFLIX TONIGHT. Spend 2 hour with this. Claude AI FULL COURSE that teaches you how to BUILD and AUTOMATE anything. The people who watch this tonight will wake up tomorrow with a new skill. Watch it and bookmark it now.
Chawanan? retweeted
whoever leaked this is f*cking crazy
someone mapped 300+ AI agents into one GitHub repo - and basically leaked the shortlist of what's actually worth trying.
coding agents. browser agents. memory. research. multi-agent teams. sandboxes.
all sitting on one page.
but everyone opens it and installs the same five names. the interesting stuff is what nobody reposts.
want an AI that hacks your own app before someone else does?
-> Strix
github.com/usestrix/strix
want a thousand agents arguing about what happens next, so you get a forecast instead of a guess?
-> MiroFish
github.com/666ghj/MiroFish
want your agent to know everything you did on your computer this week?
-> screenpipe
github.com/screenpipe/screen…
want memory that understands how facts connect, not a pile of old chats?
-> Graphiti
github.com/getzep/graphiti
want an agent clicking through a whole desktop that isn't your laptop?
-> Cua
github.com/trycua/cua
and that's barely scratching it.
~225k stars across five repos most people have never opened.
the real cheat code isn't downloading all of them.
it's opening the repo before you build anything and asking:
"has someone already solved this?"
because there's a pretty good chance the answer is yes - and the open-source version already has thousands of people testing it for you.
I went through the map and pulled out the agents I'd actually start with.
full AI agent list below
Chawanan? retweeted
One of my friend makes $1.2 million a year as an Anthropic engineer.
I asked him how he learned prompting so well. He sent me a video that was never supposed to get out. Their core team's prompting playbook.
You won’t find anything better about prompting than this 30 minutes video.
I watched it last night. Halfway through, I realized I've been using Claude completely wrong for two years.
Watch it, then read the article below.
Chawanan? retweeted
介绍比Jev快50倍,在你设备上跑的laya-mlx!
只在你的设备上占用最高1G内存
Laya是一个开源的类似于Jev的,基于文本输出概率的分类系统
我将其移植到MLX,并且做了一些性能优化!
视频中就是这个模型在我的本地M3Max上玩贪吃蛇
这个模型能够以每秒决策60次的速度玩贪吃蛇!
github.com/mizorewww/laya-ml…
- 3. Read the landscape สลับดูความลึกน้ำท่วม (Flood Depth), ระดับความสูงพื้นดินจริง (Elevation), และความลึกพื้นทะเลอ่าวไทย (Seabed Depth / Bathymetry)
📷 Give it a try here: aqshi.ur.dog














