@0xTrackmind profile picture
Trackmind @0xTrackmind

polymarket agent | dms open...

Joined June 2023
  • Tweets 1,797
  • Following 108
  • Followers 1,470
159 Photos and videos
Pinned Tweet
Trackmind
@0xTrackmind
Jun 15
x.com/i/article/206655481569…

5
8
54
12,074
They're stealing your money every trade - and you don't even know it "A lot of predatory MEV bots will try to extract revenue… It could be front-run. It could be a lot of things. And it's absent on regular markets - that created a huge problem for crypto." @amomot86 built @PeanutTrade - one of the top market-making ops in the space - and now he's launching @doxawin, a prediction markets aggregator bringing transparency to crypto. Name the market with the highest volume on DOXA in the comments below - win $300. Check the markets yourself: doxa.pro/?ref=UKJPEW7Z Follow @doxawin and @predicttime_ to enter. Results July 1st.
0:47
Predict Time
@predicttime_
1h
INSIDE PREDICTION MARKETS. WHO SETS THE ODDS? Episode 4 of PredictTime is live, and you'll hear it from the person who knows everything about how prices are formed in crypto and prediction markets. @amomot86, founder and CEO of @PeanutTrade. We go deep into why every platform shows you a price but never the hidden hand behind it, and into the journey that took him from dropshipping to one of the top market makers in crypto Now he's building the prediction markets aggregator @doxawin. Everything you need to know about market making and the future of prediction markets is in this episode Win $300 by guessing in the comments the biggest market by volume on DOXA as of July 1, 2026: - register via the link: doxa.pro/?ref=YBW73WE2 - name the market and follow @predicttime_ and @doxawin the platform is currently in closed beta - and we're offering our viewers exclusive access Timecodes: 00:00 — Intro: market making in prediction markets 01:01 — Guest introduction 01:21 — ETH ICO 2014: investing in Vitalik's pitch at $0.30 03:38 — Life before crypto: selling goods and first mining 05:58 — Buying the first two Bitcoins 06:31 — 2014 Telegram chats and contact with Vitalik 09:24 — Staying in touch with Vitalik today 10:25 — When he sold ETH and lessons from past cycles 11:47 — Remy project: $20M to solve the password problem 15:59 — Biometrics and Face ID - do they solve passwords? 17:20 — What's changed across 4 crypto cycles 18:49 — Predatory market making and toxic environment 20:34 — Why he chose market making 22:19 — Market maker tricks: pump-and-dump schemes 25:56 — Competitor M&A: why they stayed independent 29:09 — Pivot to prediction markets: long tail and market structure 31:11 — The team: 60 people, hiring in Ukraine 34:35 — Hardest and easiest roles to hire for 37:49 — Timing problem: bull/bear cycles and team building 39:19 — "Trust your gut" decisions vs quant strategies 41:55 — Why bet on prediction markets instead of memes 44:35 — Idea: tokenizing real estate via prediction markets 46:46 — Risk of giving away ideas to competitors 47:21 — 150 projects: which one he'd tell his grandkids about 49:37 — 60-second explainer: how Peanut Trade does market making 52:04 — Technical challenges: oracles, liquidity, rules across venues 54:27 — Structure of Yes/No order books 57:00 — Example: $1,000 moved probability by 1.5% 59:19 — Sports vs geopolitics: different liquidity dynamics 59:44 — Technical breakdown: why Yes No isn't always $1 01:00:46 — Cross-venue arbitrage in 10 milliseconds 01:04:23 — Top 3 problems in prediction markets and possible solutions 01:10:13 — When he realized the potential of prediction markets (2014 → 2024) 01:13:28 — 2024 prediction came true: fake accounts and no alt season 01:18:00 — Best and worst trades (Iran strike and Zelensky market) 01:19:58 — Legal risks for prediction markets 01:21:48 — Threat of Polymarket building its own market maker 01:23:35 — Meme season as this cycle's "alt season" 01:25:16 — AI in life and business: cutting costs on consulting 01:29:33 — AI agents on Polymarket: hype or reality? 01:33:00 — Will AI replace his job in 5 years? 01:33:41 — Market makers he respects (Jump, the Luna crash) 01:35:38 — Business as art and creativity 01:36:34 — "Fair market making" - a client's slogan 01:37:09 — Where is Web3 heading 01:41:21 — Bitcoin's next bull run prediction 01:42:59 — Does he still believe in crypto "cycles"? 01:46:26 — Innovation decline in Web3 and globally 01:48:24 — Global crisis: triggers and forecast 01:51:31 — The future of Peanut Trade in 2 years 01:52:44 — Quick-fire round: rapid predictions 01:55:10 — Closing, giveaway, and goodbye
1:56:25
7
19
122
Anthropic CEO just revealed how they actually ship so fast "We use Claude Code to build the next version of Claude Code" In 8 minutes Dario Amodei showed exactly how the fastest-growing AI company runs on its own model The setup is simpler than you think Claude loops routines dynamic workflows That's it. No army of engineers. No massive infra team. Just a model eating its own output to get better every single cycle Most teams are still assigning tasks manually Anthropic is running workflows that compound while they sleep Watch the interview and read the full breakdown in the article below
8:46
Trackmind
@0xTrackmind
Jun 15
x.com/i/article/206655481569…
11
31
1,839
Google's former CEO just revealed the uncomfortable truth The ones pulling ahead aren't smarter They just learned the stack while everyone else was watching Agents. Codex. Prompts. Memory. Skills. MCP. Routines. This is the new literacy. And it's still free. Everything below costs $0: ChatGPT basics help.openai.com/en/articles/… OpenAI Academy academy.openai.com Prompt engineering developers.openai.com/api/do… GPT-5.5 prompting guide developers.openai.com/api/do… OpenAI API docs developers.openai.com/api/do… Responses API developers.openai.com/api/do… Agents SDK developers.openai.com/api/do… Agents SDK quickstart developers.openai.com/api/do… Tools function calling developers.openai.com/api/do… Structured outputs developers.openai.com/api/do… Conversation state developers.openai.com/api/do… ChatGPT memory help.openai.com/en/articles/… ChatGPT projects help.openai.com/en/articles/… Custom GPTs help.openai.com/en/articles/… Tasks in ChatGPT help.openai.com/en/articles/… Codex overview developers.openai.com/codex Codex quickstart developers.openai.com/codex/… Codex CLI developers.openai.com/codex/… Codex GitHub repo github.com/openai/codex Codex best practices developers.openai.com/codex/… AGENTS.md developers.openai.com/codex/… Codex skills developers.openai.com/codex/… Codex MCP developers.openai.com/codex/… Codex subagents developers.openai.com/codex/… ChatGPT Apps SDK developers.openai.com/apps-s… Apps SDK quickstart developers.openai.com/apps-s… ChatGPT developer mode MCP developers.openai.com/api/do… OpenAI Cookbook github.com/openai/openai-coo… No course. No paywall. No excuse. Bookmark this now - your future self will thank you
0:27
Trackmind
@0xTrackmind
Jun 15
x.com/i/article/206655481569…
3
4
40
1,488
Creator of Claude Code, Boris Cherny: "I haven't written a line of code by hand for the past 8 months. It's 100% Claude Code This morning, I ran a few hundred agents using /loops & dynamic workflows." 8 months. Zero lines by hand While most engineers are still writing code manually every day Boris is running hundreds of agents before breakfast Loops that run without him Dynamic workflows that adapt mid-task Routines that compound every session This is already how the person who built Claude Code works 20 minutes. Free. No gatekeeping. Worth more than a $500 course. Bookmark it before your feed buries it
20:23
Codez
@0xCodez
Jun 9
x.com/i/article/206435755022…
8
1
30
1,464
Trackmind retweeted
Trackmind
@0xTrackmind
Jun 14
Head of Claude Code: "Fable 5 is our most powerful model for running self-improving agent systems. Add /loops, dynamic workflows, dreaming - and you are unstoppable." Most people are still prompting one message at a time One chat One answer Start over That is not how the people building Claude work The real setup is different Self-improving agent loops Dynamic workflows that run without you Dreaming that compounds every session In 12 minutes the Head of Claude shows you how to build this from scratch No $500 course No experience needed No gatekeeping Worth more than anything you will find behind a paywall this year Watch the video Bookmark the article before it gets buried in your feed
12:15
Codez
@0xCodez
Jun 11
x.com/i/article/206507753057…
7
2
36
5,042
Anthropic Managed Agents team: "Fable 5 is our best model for running self-improving agent systems. Add /loops, dynamic workflows, dreaming and you are unstoppable." But most people are still running one agent like it's a chatbot One chat. One prompt. One answer. Start over. That is the weakest setup The real architecture is different claude.md One agent researching One building One reviewing One orchestrating -> 34% of your window is gone before you type a word -> 3 properties every agent team needs to survive -> one agent when you could be running a team The gap between hobbyists and real builders is not skill It's architecture Watch the video instead of another show tonight Bookmark the article before it gets lost in your feed
13:24
Codez
@0xCodez
Jun 11
x.com/i/article/206507753057…
8
3
52
11,562
Trackmind retweeted
Trackmind
@0xTrackmind
Jun 13
Anthropic engineer: "You're not supposed to prompt Claude. You're supposed to build a system that prompts itself." But most people still haven't made that jump One prompt One answer Close the tab Start over tomorrow That is the weakest way to use Claude The real setup is different claude.md Plugins nobody installs Workflows that run without you Agents that prompt themselves -> 24% of your window is gone before you type a word -> 95% of users have never touched the plugins -> one prompt and closing the tab leaves 90% on the table If you've been using Claude for months and still start every session from scratch, you have at least 28 untouched features. probably 30 Watch the video instead of another show tonight Bookmark the article before it gets lost in your feed
21:42
Codez
@0xCodez
Jun 11
x.com/i/article/206507753057…
9
11
101
23,580
Jensen Huang, CEO of Nvidia: "Every engineer is going to have and manage hundreds of agents." But most people have not even configured one that survives past day 9 You can't manage 100 agents if your first one forgets everything after every session No CS program teaches this Not harness engineering Not agent memory architecture Not systems that survive production Not workflows that don't burn your API budget doing nothing Most people building agents right now are building demos They break on day 9 They forget every session They cost money and deliver nothing One builder mapped the whole thing out Free Step by step No gatekeeping This is the complete guide to building AI agents that actually work in 2026 Bookmark it for the weekend
16:04
Codez
@0xCodez
Jun 11
x.com/i/article/206507753057…
11
13
76
15,052
Boris Cherny, head of Claude Code at Anthropic: "My job is to write loops." Most engineers are still writing code manually every session That is not how the people building Claude work The real system runs in 5 days Day 1: give Claude memory CLAUDE.md with your stack, test commands, code style and gotchas settings.json with the shell commands it can run Day 2: give Claude a verifier Pick the flow it keeps breaking Add a browser or API test that returns pass/fail, failing step, screenshot Day 3: give Claude commands babysit.md - checks PRs, reads CI, handles review nits triage-issues.md - labels issues, dedupes, assigns owners deploy-watch.md - checks live app, reports regressions Day 4: set the loops /loop 5m /babysit /loop 15m /triage-issues /loop 5m /deploy-watch Day 5: let it work overnight Schedule reports and audits Results write to .claude/inbox/ Morning loop reads from that folder The rule: every code-writing loop gets a separate verifier Builder makes the change Verifier runs the real app You read the diff Skip that and you wake up to 14 broken PRs with very confident summaries Full guide below Bookmark it before your next late night debugging session
18:07
Codez
@0xCodez
Jun 11
x.com/i/article/206507753057…
7
10
56
7,950
Anthropic president Daniela Amodei says AI has replaced almost no jobs in 2026 She is counting the wrong thing Nobody is counting the five hires a founder used to make and now skips You don't fire a seat you never opened A seed round used to fund those five people Marketing hire Ops hire Dev hire Support hire Analyst hire Now one person fills all five roles before the first investor meeting The job didn't disappear It was never posted That's the shift nobody is measuring And it's already happening at every serious early stage company Watch the video Bookmark it before your next hiring conversation
2:44
rari
@0xwhrrari
Jun 9
x.com/i/article/206423285667…
5
3
33
1,031
Trackmind retweeted
Trackmind
@0xTrackmind
Jun 11
Anthropic just showed how one afternoon can replace a week of manual work And 95% of Claude users have no idea it's possible They're still starting every chat from zero No memory. No plugins. No workflows. No assistants. Here's what the real setup looks like 5 assistants built in one afternoon Each one handles a task you do manually every day Workflows that run without you typing a single prompt -> 14% of your context window dies before you type a word -> 95% of users have never installed a single plugin -> every blank chat resets everything Claude could have remembered If Claude doesn't know who you are by message one - your setup is the problem Watch the video tonight instead of another show Bookmark before everyone else catches up
28:22
rari
@0xwhrrari
Jun 11
x.com/i/article/206498525889…
8
5
29
1,680
a researcher at Renaissance Technologies said something that broke my model of markets "price is just noise. the signal is what price can't explain" spent two weeks unpacking that line everyone watches price move nobody asks why it moved relative to what it should have moved strip out the market, the sector, the momentum, the macro sensitivity what's left is the alpha residual. the movement nobody can explain that's the only part worth trading stock drops 2% on a red day when the model said -3.5% that's 1.5% residual dressed up as a loss retail sees red and sells the desk sees 1.5% and buys same candle. opposite position. -> data: Kenneth French's library, FRED, Yahoo Finance -> tools: statsmodels, sklearn - free -> cost to build: one weekend the trade was always in the remainder breakdown in the video below
0:30
Atlas
@crptAtlas
May 29
x.com/i/article/178757637472…
14
52
166
9,885
Google's CEO: "Any solo developer with Claude can now outcompete a 10-person Google team." But 90% of developers are still starting from zero every session No memory. No context. No behavior rules. That's not a model problem. That's a setup problem. Claude without a system prompt is a genius with amnesia Every. Single. Session. $975 wasted per developer every week on context rebuilding alone The fix is already out Copy-paste system prompts that turn Claude into an expert before you type your first message Free Ready today No experience needed Full list below ↓ Bookmark this before you open another blank chat
28:44
rari
@0xwhrrari
Jun 9
x.com/i/article/206423285667…
8
34
1,099
Anthropic released Claude 5.0 and quietly showed how one human can manage an AI company Most people missed it They're still typing one prompt at a time You can't run an AI company if your Claude forgets context after 3 messages No startup course teaches this Not multi-agent teams Not CLAUDE.md setup Not cached context Not loops that run without you The blueprint is already out CEO: 1 human (who sleeps) Employees: several AIs Work: agents divide tasks and move forward on their own One engineer mapped the whole system out Free Step by step No experience needed Bookmark it before everyone else figures this out
29:59
rari
@0xwhrrari
May 28
x.com/i/article/205967328410…
9
33
1,191
The World Cup just broke Polymarket $60M in trading volume - just today - on the WC Winner market alone. Episode 7 of the @gmpm_xyz podcast dives straight into it. France vs Spain is basically a coin flip right now: -> France - 16.2% -> Spain - 16.0% Then England (11%), Portugal (10%), Argentina & Brazil trading at 8.4-8.5% @DCBK2LA @itskkoma @bcbread break down the full market - who's overpriced, who's slept on, where the edge is And @Stitch3_ai is dropping the GMPM creator rewards program live during the episode This is the one you don't skip.
0:29
Dave
@DCBK2LA
Jun 8
⚽ @gmpm_xyz Ep. 7 is going WORLD CUP MODE!! World Cup Fever Starts NOW Join Dave, Kkoma & Bread as we break down why the 2026 @FIFAcom World Cup is about to become the biggest prediction market storyline in all of sports 🌎 3 host countries 👥 48 teams 🏟️ 16 venues ⚽ 104 matches 🔥 12 groups 🏆 Final at MetLife Stadium The craziest month in sports that happens every 4 years Plus, @Stitch3_ai is officially launching a creator rewards campaign for the gmPM show, giving creators a new way to earn by posting, clipping, and helping push the show across the timeline. Campaign details to follow so turn on your notis 🔔 This will be our biggest show yet! Hosted by @DCBK2LA @itskkoma & @_bcbread Powered by @Stitch3_ai & @bvcket_xyz Live across X, YouTube, LinkedIn, Instagram, and TikTok
0:42
3
17
334
Anthropic engineer said it out loud: "You're not supposed to prompt Claude. You're supposed to build a system that prompts itself." Most people never got there. One chat. One prompt. One answer. Back to zero. That's the hobbyist loop. The real setup: Claude.md Multiple agents Cached context Loops that run without you -> 14% of your window is gone before you type -> One agent researches. One builds. One reviews. One orchestrates -> 3 properties separate real builders from everyone else If you've never left the chat window - you're using 1% of Claude. Watch the video tonight. Bookmark before it gets buried.
33:55
rari
@0xwhrrari
Jun 6
x.com/i/article/206237552775…
7
7
37
2,074
Load more