Don

@bored2boar yeahhhh slide me this backtest data bruh

Your Polymarket bot is failing for one simple reason: Your backtest is USELESS. And it’s not because your strategy is bad. It’s because your data is incomplete. I built 1,000+ bots and 99% of them failed. But this 1% still prints for me DAILY (over $16k public PnL this month). Most people test using APIs, Binance data, or simplified price feeds and think that’s enough. It’s NOT even close. You’re missing order book depth, real spreads, and most importantly actual fill behavior. That’s why your bot looks profitable on paper but breaks the moment you go live. You’re not simulating reality. Fix is simple, but almost nobody does it. Build a recorder and start now. Track every tick of every market you trade, including price changes and full order book depth. Also record every external data source you rely on like NOAA, ESPN, Binance, or Coinbase. And don’t just track the active window. You need pre-window and post-window data too, because timing is where most of the edge comes from. The biggest mistake is fills. If your model assumes perfect execution, it’s already wrong. That’s exactly why people complain about getting only 40% fills live. They never tested it properly. You need scale. At least 1000 windows for statistical confidence, which is around 10 days on 15m markets. NO real data means NO real edge. Just a strategy that works only in your imagination. Preparing a full guide for you, dropping in ~24 hours.

looking for few people to test this out and give feedback making discord later today, dm/reply if interested in joining github.com/rohunvora/past…

@gmoneyNFT shoulda just stuck w openclaw and spent time dialing it tf in so it stops breaking. this is the method and everything else is just temu version of openclaw icl

i gave it a shot, but can't do this anymore. hermes sucks ass. all these agents suck ass. they just stop working all the time and then take forever to debug. sticking to claude code and codex in terminal. far and away better than messing with this productivity porn

@AlexFinn skill issue

IF YOU'RE ON OPENCLAW DO THIS NOW: I just sped up my OpenClaw by 95% with a single prompt Over the past week my claw has been unbelievably slow. Turns out the output of EVERY cron job gets loaded into context Months of cron outputs sent with every message Do this prompt now: "Check how many session files are in ~/.openclaw/agents/main/sessions/ and how big sessions.json is. If there are thousands of old cron session files bloating it, delete all the old .jsonl files except the main session, then rebuild sessions.json to only reference sessions that still exist on disk." This will delete all the session data around your cron outputs. If you do a ton of cron jobs, this is a tremendous amount of bloat that does not need to be loaded into context and is MAJORLY slowing down your Openclaw If you for some reason want to keep some of this cron session data in memory, then don't have your openclaw delete ALL of them. But for me, I have all the outputs automatically save to a Convex database anyway, so there was no reason to keep it all in context. Instantly sped up my OpenClaw from unusable to lightning quick

@0xRiver8 @esblee2 i respect it river. keep doing you g

@esblee2 how can something so hot be cold?😅😁😆😊

it’s getting cold in new york again 😭

Wait y’all are still trenching in this market….?

@AlphaStratAi @patty_fi sonnets insufficent i only use top model from each provider

@DaDefiDon @patty_fi I seen Sonnet was working

anyones agents failing rn on claude max for literally no reason?

@AlphaStratAi @patty_fi i use claude for everything and use openai as a fallback. i run any crons that arent super high level on codex tho to try slow down how fast i hit my claude max limit

@DaDefiDon @patty_fi Exactly what we did, Claude for Big projects, and OpenAi as a backup for small tweaks through the day just incase and have them feeding information back and forth like a team

@paularambles what the fuck is happening rn do agents love lo-fi

it's happening

@ashen_one just drop link tf

I finally found a website that I can confidently recommend people set up Openclaw on virtually I just shot the video for it, and it literally took me about 4 minutes to set up Openclaw I've set up 10+ Openclaws on 4 different maps, and each one has taken me a minimum of an hour If you're someone who still hasn't bought a Mac mini for Openclaw and are thinking about whether they should or shouldn't I highly recommend you watch this video before you do that and look into hosting compared to physical hardware.

@shafu0x mine has access to reddit

@melvynx well duh! openclaw is for spawning subagents or claude code/codex acp’s and prompting them to code

OpenClaw is not for coding. OpenClaw is not for coding. OpenClaw is not for coding. OpenClaw is not for coding. OpenClaw is not for coding.OpenClaw is not for coding.

@elie2222 @inboxzero_ai ok thank you was just checking if i needed this lol

@DaDefiDon @inboxzero_ai This is to manage a Gmail our Outlook. If your agent has its own email then not relevant. But if you want it to manage your personal email then it is.

Inbox Zero is live on ClawHub 😍 OpenClaw can now manage your @inboxzero_ai account. But why? Can't OpenClaw handle email? Well, there are a few reasons: 1. Claw can handle email, but no one trusts it to 2. It costs an insane amount for Claw to handle email 3. Claw isn't built for email. More of a general purpose assistant. With the OpenClaw <> Inbox Zero skill/CLI, your Claw can now manage your IZ account without putting you at heavy security risk. OpenClaw becomes the manager of your Inbox Zero agent.

@elvissun ima start doing this for claude limits 😂 seems theres no way to view codex limits that ive found tho

if you have unused weekly limits the best way to burn them is just spamming fan-out deep research in cc/codex: - 0 review cycles needed - context-dense files you reuse forever - no slop generated (it's source material, not final output) - feeds into content, product, marketing or competitor intel later ran 22 parallel research agents to burn through ~15% of weekly usage in 20 minutes. tokens very well spent.

@MidCurveMortal @chang_defi @NousResearch its dogshit its a 9b model bro

@chang_defi @NousResearch How does it compare to Anthropic / claude when it comes to reasoning skills ?

cancel your chatgpt subscription and delete your openclaw slop. i'm serious. go on ebay and buy a used RTX 3060 for the price of two months of pro. or check your drawer because half of you already own one and forgot about it. install hermes agent from @NousResearch. one framework, 31 tools, file operations, terminal, browser, code execution. connect it to your local llama.cpp server running qwen 3.5 9B Q4. total download is 5.3 gigs. that's it. that's the whole setup. every experiment you hesitated to run on API. every project you shelved because you didn't want your data on someone else's server. every late night idea you didn't test because you hit your rate limit. all of that is gone. runs 24/7 on your electricity. your machine. your data never leaves your house. connect it to telegram if you want it on your phone. hook up whatever tools you need. the model thinks at 29 tok/s with 128K context and it never bills you. qwen 3.5 9B and one RTX 3060 is the setup most people will never try because they've been trained to believe intelligence has to come from a datacenter. it doesn't. it runs on 12 gigs of VRAM under your desk right now. stop giving your thinking away for free.

@donatedotgg holy GOATs

Introducing Campaigns, a new feature for Donate. Easily create fundraisers for yourself or nonprofits and accept contributions in seconds.

@ufc @paramountplus no fucking way the carda this shit

THE OFFICIAL BOUT LIST FOR #UFCWhiteHouse [ LIVE Sunday June 14 on @ParamountPlus ]

@finnbags @BagsApp ayo whaaaa

@BagsApp this is insane 🤯

The Bags Hackathon is LIVE NOW. $4,000,000 in funding for builders on @BagsApp Apply to the Q1 2026 cohort here 👇 bags.fm/hackathon