shmidt
9.6K posts
shmidt
@shmidtqq
growing stronger with every passing day / @zscdao
➜ Entrou em Ekim 2021
900 Seguindo2.3K Seguidores
While all this quants in the timeline are printing millions with AI, you’re still lost in simulation models and deep research agents.
Don’t worry - I found solution for you. 18k stars on GitHub. Your own financial analyst right in your pocket.
An autonomous financial research agent that thinks, plans, and learns as it works - Dexter.
> 2-command install, 30 seconds to set up
> full Claude skill set + agentic search APIs
> real-time market data straight from financial datasets MCP Server
> text it on WhatsApp for research if u want (financial Jarvis living in your phone)
Just paste GitHub link into ChatGPT/Claude/Grok or any other LLM and it will walk you through the two-click install.
Everything else is handled by Dexter.
It performs analysis using task planning, self-reflection, and real-time market data.
Code is 100% open source and available on GitHub:
github.com/virattt/dexter
Save this post so you don't lose the links.
cvxv666@antpalkin
Chinese quant built a simulation of how SPX price reacts to any global event. He’s already made over $100k - with full blockchain proof. He knows exactly where price will go. More than 40 years of SPX trading history have been loaded into MiroFish simulator (18k stars on GitHub) AI analyzed every single moment in that trading history. Now this guy has a fully functional SPX price prediction system. His wallet: @moisturizer?via=cvxv666" target="_blank" rel="nofollow noopener">polymarket.com/@moisturizer?v…
Dozens of successful SPX price-prediction trades and hundreds of tests across other stock markets. Here’s exactly what you need to replicate his stack: - market data APIs (SPX price, use Alpha Vantage or Quandl) - data pipeline (use Python) - feature engineering (for output signals like RSI, MACD) - seed dataset for MiroFish (convert data into structured context) - multi-agent simulation (macro strategist, earnings analyst, sentiment analyst agents etc.) - probability forecast (run different scenarios) - trading / decision Model (SPX futures ES, SPY ETF) Save this pipeline if you want to run a similar simulation on your own data. You can feed the whole thing to your Claude and build your first (even small) simulation model together. English
how does it feel having your own pocket financial AI agent that instantly cooks up any market analysis or full news recap the second you shoot it a WhatsApp text
cvxv666@antpalkin
While all this quants in the timeline are printing millions with AI, you’re still lost in simulation models and deep research agents. Don’t worry - I found solution for you. 18k stars on GitHub. Your own financial analyst right in your pocket. An autonomous financial research agent that thinks, plans, and learns as it works - Dexter. > 2-command install, 30 seconds to set up > full Claude skill set + agentic search APIs > real-time market data straight from financial datasets MCP Server > text it on WhatsApp for research if u want (financial Jarvis living in your phone) Just paste GitHub link into ChatGPT/Claude/Grok or any other LLM and it will walk you through the two-click install. Everything else is handled by Dexter. It performs analysis using task planning, self-reflection, and real-time market data. Code is 100% open source and available on GitHub: github.com/virattt/dexter Save this post so you don't lose the links.
English
the way my portfolio looks at me now after one weekend with MiroFish swarm
> 2 years heemed on Polymarket
> Chinese undergrad drops AI swarm
> now printing like quants with the full matrix and real-time alerts from polyalerthub.com/?via=slash1a486
retail cooked.
slash1s@slash1sol
Mind blown 2.0: Onchain quants just printed $400,000+ trading Polymarket bets on SPX, Dow, Russell 2000, AAPL, GOOG - all powered by open-source financial market simulations! The god-tier stack just dropped: Financial Datasets MCP Server (1.7k stars) + MiroThinker-H1 (88.2 benchmark, 7.1k stars) + MiroFish - multi-agent simulation engine built by a Chinese undergrad student in just 10 days (now 18k+ stars and scored millions in funding overnight). How each repo works and how to apply it: 1. Financial Datasets MCP Server -> Unlimited real historical prices, balance sheets, income statements and news for ANY ticker (SPX, AAPL, GOOG, Russell 2000 etc.). This is your free 40+ year data parser - just connect and pull raw facts. Repo: github.com/financial-data… 2. MiroThinker-H1 -> Deep research agent (pulls data straight from the MCP Server). Analyzes latest 10-K, Fed minutes, earnings, geopolitics and builds 500+ clean, verified datasets. Without it your simulation is garbage - it turns raw data into the perfect “brain” for the engine. Repo: github.com/MiroMindAI/Mir… 3. MiroFish -> The actual multi-agent simulation engine. Load the dataset from MiroThinker and run thousands of AI agents with different personalities (macro strategist, sentiment analyst, panic buyer etc.). Get probability cones and the full "matrix" - exactly how price will react to any event. Repo: github.com/666ghj/MiroFish Key Applications: .Trading: throw in any event -> simulate crowd reaction -> catch the edge and ape on Polymarket .Macro forecasting: test global events before they hit the news .Easy setup: Docker + any LLM API, live in 10-15 minutes Pro tip: Feed MiroThinker latest 10-K or breaking news -> it builds 500+ verified scenario datasets -> load into MiroFish -> get probability cones for next-week price moves. Then ape the highest-conviction side on Polymarket risk-free. Traders are already winning big: [superstonksbro] -> PnL = $182k, multiple $20k+ wins. @superstonksbro?via=svyatoslav" target="_blank" rel="nofollow noopener">polymarket.com/@superstonksbr…
[CamelUp] -> PnL = $193k, 2.4k+ predictions. polymarket.com/profile/%40Cam… Both crushing it with this exact stack. For effortless gains, try Kreo copy-trading: auto-mirror these new simulation beasts and ride their edges. Try here: @join" target="_blank" rel="nofollow noopener">kreo.app/@join Add their wallets: 0x17559efac103ac7f361be37ec0b93888d4c55aac // 0x969fae0a3a93778adc42178f72c612ed8c4e4d55 to [t.me/KreoPolyBot?st…] and start track/copy them right now. Save this links and info so you don't lose it. English
Roma vs Bologna
Here, I also expect a repeat of the first match.
Roma are essentially the main favorites of the tournament, but Italiano and the team have shown that they won’t give up their quarterfinal spot easily.
Bologna has found good form recently, and I expect them to score at least one goal. The Gasperini team shouldn’t disappoint their fans either.
My prediction on Polymarket: both teams to score
English
I've finally found the perfect option for copy trading - 99% WIN RATE
$752,005 in total revenue. 25,574 forecasts. No interruptions since January 2025 on @Polymarket
No panic. No speculative markets. No big bets.
99% Win Rate. 25,574 deals. This isn't a fluke - it's a system.
His profile - @sharky6999" target="_blank" rel="nofollow noopener">polymarket.com/@sharky6999
For copying, I recommend using Fors: app.fors.market
Dekos@PolyDekos
$739 → $1,817,710 betting on NBA spreads This Polymarket trader entered with $739 in November 2025. No complex models, no insider information, no high-profile bets No setbacks. No panic. Just edge by edge. his profile - @gatorr?via=dekos2911" target="_blank" rel="nofollow noopener">polymarket.com/@gatorr?via=de…
English
Mom: what do you do for work
Me: I manage three AI agents that made $30K last month
Mom: are they single
zostaff@zostaff
English
His name is bin8888. He integrated his own feed parser directly into his terminal and earned $58,354.54 in 72 hours.
While retail investors were nervous about delays in macroeconomic news, he launched a Rust-based WebSocket parser on the command line to retrieve OPEC+ insider leaks in fractions of a second.
bin8888 set up a lightweight empirical data processing pipeline to detect changes in the probability of oil supply shocks, converting raw geopolitical chatter into pure alpha returns.
When his terminal generated a hidden consensus change signaling stabilization of oil production, he reacted instantly.
bin8888 aggressively attacked the "No" side of Polymarket's March crude oil contracts.
Having abandoned strike prices of $110, $120, and $105, he bought over 240,000 shares for pennies while the market was still pricing in the supply crunch.
His dashboard lit up: position value $156,000, net profit $58,354.54.
If your terminal doesn't read the news before it's published, you're a source of liquidity for the exit.
English
I SAW THIS ARTICLE AT 11:27 PM AND DIDN’T SLEEP UNTIL 4:11 AM
Read it three times, then just... started building.
Took the Avellaneda-Stoikov quoting logic.
Wired it to the Hawkes process for order flow.
Added the VPIN circuit breaker exactly like the article says.
Ran 500 simulations tonight:
> Mean P&L: +$312/session
> Sharpe: 1.87
> Win rate: 68%
> VPIN saves: 11 sessions
> Max drawdown: -$890
The VPIN part is insane btw, 11 times it pulled my quotes before informed flow ran me over, without it sharpe goes negative, just like that.
Kyle's lambda estimation is literally 30 lines of python, i had no excuse not to build this.
Side effect, now i can't sleep.
verax@journoverax
English
@slash1sol @PolymarketTrade saved
fast and clean data is the right key to success
English
Want to feed your bots clean, high-quality market data so they actually print on Polymarket instead of guessing?
Part 3 of the 12-part series just dropped.
Now it shows exactly how to:
-> Connect to Binance API and pull historical candles
-> Clean & process everything with pandas
-> Calculate simple + log returns
-> Visualize charts + moving averages
-> Save data fast (CSV + HDF5) with ready-to-use data_loader.py
This is the real fuel. No good data = dead strategy.
Real proof this method works right now:
One Python bot..
[LucasMeow]
..has already printed $277,186 all-time profit on Polymarket.
- $716.3K Positions Value
- Biggest single win: $26.8K
- 148 predictions (crushing SOL dips to $140/$80 and ETH targets)
Profile -> polymarket.com/profile/%40Luc…
Want this exact type of edge running automatically without writing any code yourself?
Kreo automatically copies top-performing bots like this one instantly, set it and forget it.
Try here: @join" target="_blank" rel="nofollow noopener">kreo.app/@join
Add his wallet 0x7f3c8979d0afa00007bae4747d5347122af05613 to [t.me/KreoPolyBot?st…] and start track/copy him right now.
Save this.
sopersone@sopersone
English
English
Most people think Polymarket runs on LMSR - where liquidity is always supplied by math!
In reality, the platform has fully switched to CLOB (Central Limit Order Book)
This isn't just a tech upgrade - it's a full shift from algorithm to participant-driven market.
What changed:
＞Before: liquidity came from a formula → you could always trade, but as volume grew, costs and slippage shot up
＞Now: liquidity comes from traders and pro market makers → prices are set by real order flow in the book, just like on a regular exchange
Key advantages:
＞Capital is way more efficient - you no longer have to lock it across the entire 0-100% range
＞Algo market makers keep tight spreads 24/7
＞Polymarket is hybrid: off-chain matching (speed) and on-chain settlement on Polygon
＞Liquidity Rewards and Maker Rebates replaced the old hidden LMSR subsidy
＞Risk is now distributed among participants
Polymarket evolved from a convenient math model into a real prediction exchange with actual strategies: arbitrage, liquidity providing, and order book dynamics
Great breakdown by @polyfactual - must read
Polyfactual@polyfactual
English
Crypto hate at absolute peak. Launch delays, project shutdowns. Vanity Fair dragging us like we’re literal clowns. Whole timeline pivot to AI.
I’m turning this backlash into $10k -> $60k single trade with ridiculously low risk.
Thanks to @polyboostxyz for the cheat code - bundle every undervalued Polymarket outcome into one parlay.
600% ROI just because these projects are finally launching.
Largest crypto wallet ever at $300M FDV.
Largest NFT marketplace ever at $100M FDV.
Just a few months ago everyone was valuing these things in the BILLIONS.
Today the market is handing us 6x simply for them to launch trading at all.
At least at Tier-3 crypto project prices.
I literally cannot imagine a universe where they’re worth even a penny less.
This is the most insane market inefficiency I’ve ever seen - and I closed it in two clicks with a parlay by this guys:
polyboost.xyz/?ref=cvxv666
Be sure to check it out - Polyboost also features a powerful AI betting assistant, 60-second markets and various bonuses.
Access is by invitation only, only 25 FCFS spots available via my link.
English
Olympique Lyonnais vs Celta Vigo
The first leg ended in a draw, so neither team has an advantage going into this match.
Olympique Lyonnais are currently on a poor run, winless in their last 6 games after a streak of 13 consecutive victories. I expect that run to end today.
Celta Vigo are having a strong season, but for some reason I don’t believe they will make it to the next round.
Even in the first match, Olympique Lyonnais looked more confident and created more chances.
I expect Lyon to finish the job and get the result.
My pick on Polymarket: Lyon to win
English
🇱🇧🇮🇱 Lebanon's death toll from the war with Israel and Hezbollah has surpassed 1,000.
118 children. 79 women. 2,584 wounded. Over one million displaced.
Lebanon has seen this story before…
Source: AP
Mario Nawfal@MarioNawfal
🇮🇷🇶🇦 Qatar Prime Minister on Iran attack: Says damage can be fixed, lives cannot Calls strike on key economic facility unacceptable Says it harms vulnerable people worldwide, hopes Iran understands
English
🚨 NEW POLYMARKET: What will Trump say during events with Japanese Prime Minster? polymarket.com/event/what-wil…
English
@Atenov_D @Polymarket @PolymarketTrade @zscdao @GreatUGB @KrRyMa @Yarpin5 @RAYYAN_THE_OG @sharmabi86 @Cattobreed @sflorimm i saved my 200 hours, thanks for the summary
English
Matthew Berman spent 200 hours perfecting his AI agent. Here's what he learned:
> Most people set up an agent once and wonder why it underperforms. The gap isnt the model. Its the architecture around it.
- Never use one long chat.
Create a Telegram group - just you and the bot and enable Threads. Separate contexts: General, Knowledge Base, CRM.
The agent loads only what's relevant, stays focused, works faster. One long chat is a context graveyard.
Voice messages work perfectly for long tasks on the go. Use them.
- Stop paying per token.
API costs compound fast with an active agent. Integrate via Agents SDK (Anthropic) or Codex Auth (OpenAI) to use your existing consumer subscriptions at a fixed monthly rate.
- One model for everything is wasteful.
Route by task. Sonnet/Opus for planning and core conversations. GPT-5.4 as fallback. Gemini for search and video.
Free local models like Qwen for email sorting and routine triage. Pin specific models to specific Telegram threads.
Anything taking longer than 10 seconds - code, API calls, file work - gets delegated to a sub-agent automatically. Your main agent never blocks.
- Claude and GPT need different prompts.
Claude hates CAPS LOCK and "dont do X" instructions. GPT is the opposite. Keep separate system prompt files per model.
Run a nightly cron to sync them by meaning while preserving each model's ideal formatting.
- Move heavy tasks to 3am.
Backups, code checks, documentation updates - all scheduled overnight. Your daytime quota stays intact for actual work.
- Security is non negotiable.
Filter all inputs through a code layer first, then a separate LLM that quarantines suspicious content.
Strip PII from all outputs automatically. Set hard spending limits - an agent caught in an error loop will drain your budget in minutes.
Read-only permissions by default. Manual confirmation for anything destructive.
- Notifications will drive you insane without batching.
Non-critical alerts: one digest every 3 hours. Medium priority: hourly. Instant delivery only for critical failures.
200 hours of mistakes, compressed into 8 rules.
Bookmark this. A few hours to set up. Compounds for years.
Atenov int.@Atenov_D
I asked my AI to research a topic. It returned confident nonsense. Here's why and how I fixed it in one folder. Standard AI memory is a black box. The agent saves data as vectors - numbers you cant read, cant audit, cant trust. > Ask it to research something and it returns polished nothing. AI slop with emotional flourishes instead of actual analysis. The fix is embarrassingly simple: save everything as Markdown files in Obsidian, backed up to GitHub. Transparent, readable, yours. By week 4-5 something shifts. The agent stops writing like a robot and starts writing in your voice. It connects ideas proactively - things you never explicitly linked. > Thats not a feature you turn on. Its what happens when the memory has structure. Morning intelligence pipeline. Dont use one prompt to gather all your news. Split the work across parallel sub-agents - each one focused, each one faster. Set a cron job for every morning. Specify your local timezone or the agent runs on UTC and your briefing arrives at 3am. The agent scans via Brave Search, X, financial sources - pulls the top 10 events, writes a 2-minute summary directly into Obsidian. If anything touches an investment idea or content angle, it adds it automatically to your ideas backlog file. AI as a trading assistant - and the truth about @Polymarket Arbitrage bots dont work. Hidden platform fees kill the math. Ignore those threads. What does work: give the agent your actual trade journal. Wins, losses, dates, conditions. Ask it to find the hidden variables - what's consistently present in your profitable trades and absent in the losers. Order flow, delta, volume patterns. The agent spots correlations in seconds that would take you weeks to see manually. Your edge already exists in your data. The agent just reads it. > Two plugins that make the architecture work. Smart Connections - local vector search, no API costs. Connects ideas across your vault automatically. QMD as MD - forces the agent to save all outputs in clean Markdown. Obsidian reads everything correctly. Readable memory beats black box memory. Every time. Bookmark this. A few hours to set up. Compounds for years.
English
@PolyDekos he's just starting his career, we need to keep an eye on him
English
$162,906.72 in 3 days on Polymarket sports markets
No crypto. No politics. No distractions.
9 predictions. Positions worth $173,900. Largest win: $35,500.
It’s not the number of deals that matters - it’s the quality of each lead.
His profile - @Oliverdjob?via=OsnO" target="_blank" rel="nofollow noopener">polymarket.com/@Oliverdjob?vi…
English
English
English
This guy bet $1,000,000 on Middle East geopolitics and his predictions are frighteningly accurate.
Meet the trader known as trevors4.
Some of his statistics:
-> Total profit: +$83,634 in just a couple of weeks!
-> Win Rate: 88.9% 🤯
-> Current Portfolio Value: $981,170
-> Predictions: 14
But the most interesting part is WHAT he's making this crazy money on. He's not trading Bitcoin or Sport makrets. He's trading war, peace, and the fate of nations. And judging by his $1M portfolio, he knows something they aren't showing on the news.
Here is his plan for the near future:
❌ $260,000 that there will NOT be a US/Iran ceasefire by March 31.
❌ $198,000 that the Iranian regime will NOT fall by March 31.
❌ $184,000 that Netanyahu will STAY in power through March.
❌ $169,000 that US forces will NOT enter Iran by the end of spring.
I don't know what will happen tomorrow, but I'll definitely be keeping an eye on every dollar this trader moves. If he starts closing or liquidating positions abruptly, it means something is brewing.
🔗 His profile: @trevors4?tab=positions&via=all" target="_blank" rel="nofollow noopener">polymarket.com/@trevors4?tab=…
English
@shmidtqq @zscdao @Polymarket @PolymarketTrade I think it's a reasonable decision to track it for now.
English
English