WeGo2Mars

43. Self-taught. Building an AI trading company in public. Gold is the proving ground. 4 AI agents. Real backtests. Real mistakes. We kill weak strategies and keep receipts. If you want guru fantasy, wrong account. If you want the build in real time, follow along.

Most people want financial freedom. Very few are willing to spend 2 years learning a market, building a system, and watching it fail before it works. The ones who do are not special. They just did not stop. #trading #discipline

@adamwathan For trading systems this is brutal. Name the specific market condition your strategy was built for. If you cannot, you do not have a strategy yet. You have a backtest. The real user is yourself after 6 months of live exposure.

If you're building a new product, make sure you can name a real human being who you think would be a customer. Not a persona, a person. It sounds so extremely obvious but it's so easy to get lost in the details of a "good idea" and stop evaluating it against this basic test.

@ChrisLaubAI The framework-agnostic angle is underrated. We spend more time fighting framework complexity than building the actual agent logic. Anything that reduces that friction is worth testing. Adding this to our pipeline this week.

🚨 This project just made most AI agents look outdated. It’s called Better Agents and it supercharges your coding assistant (Kilocode, Claude Code, Cursor, etc), making it an expert in any agent framework you choose (Agno, Mastra, etc) and all their best practices. This is the future of autonomous AI. Here’s how it works 👇

@VittoStack The multi-agent coordination point is the one that matters most for trading. Single agent for a single task is dead. The edge is in specialized agents each owning a piece of the pipeline. That is the direction we are building. #AI#AI

Anthropic just released its 2026 Agentic Coding report. If you're building with AI agents, bookmark this ✨ It covers 8 trends reshaping dev: - single agents → coordinated multi-agent teams - autonomous builds running for days, not minutes - agents that know when to ask for help - non-technical teams building tools without engineers - security scaling on both sides of the fence Software development is shifting from writing code to orchestrating agents that write code.

@freedombill3 Same applies to trading systems. Manual execution is the enemy of consistency. Once you automate the rules, you remove yourself from the equation. That is when the math can actually work. #trading#trading

When you automate your investments, you stop relying on motivation and start relying on mathematics. 1p Challenge Day 69. I have moved 69p in to the $VHYL, $VUSA and $JEPQ pies. Building wealth requires constant discipline, which is why most people eventually stop. These daily deposits, I remove emotion from the equation. Today, the machine quietly pulled another 69p into each pot, pushing the total challenge value up to £72.56. It is the proof that you do not need huge chunks of spare cash to start building a foundation. $VHYL Pot • Cash Invested: £24.15 • Current Value: £24.03 • Divs Earned: £0.00 $VUSA Pot • Cash Invested: £24.15 • Current Value: £24.16 • Divs Earned: £0.00 $JEPQ Pot • Cash Invested: £24.15 • Current Value: £24.37 • Divs Earned: £0.08 Total Income: £0.08 70p tomorrow.

@jadecap_ Trading specifically. One missed execution rule undoes months of consistency. The discipline is not in the good trades. It is in the bad setups you pass on without flinching.

If you struggle with discipline, start in one area until you can execute without deviating from the plan. Then begin to tackle other aspects. You cannot force discipline in every area right off the bat... it is overwhelming. Struggle with risk? Practice strict position sizing. Struggle with over trading? Practice limiting frequency. Struggle with emotions? Practice mindfulness. Over time you will master self-discipline.

Every trade I simulate, I ask: would I still take this if I had to explain it to someone smarter than me? Most ideas fail that test. The ones that pass are the only ones worth coding. #trading #AI

@B_trader__ Saw it and called it. Gold has been the most consistent macro trade of the decade. Every time someone says the run is over it just recycles higher. The structural bid from central banks alone justifies being long. #XAUUSD

XAUUSD Possible ATH in View on GOLD. Do you see what i see?

@Saboo_Shubham_ The real test of self-improving AI: does it improve toward what you actually want? Running this on a trading agent. Better evals do not equal more money. That gap is the whole problem. #AI#AI

Self-improving AI Agent skills using Gemini 3. Just upload your skills and watch it improve in real-time. 100% Opensource. Launching soon.

Building an AI trading system in public. The hardest part is not the code. It is deciding what is worth building at all. Most ideas look smart in a backtest. Very few survive contact with live markets. We are learning that difference right now. #gold #buildinpublic

@MARA Selling $1.1B BTC to cut 30% of convertible debt. Love the Bitcoin strategy but also respect the discipline. Most companies lever up, lever up, lever up until the margin call. MARA is actually managing risk. That's the move.

Today, MARA announced the repurchase of ~$1B in convertible notes at a ~9% discount to par value. ~30% convertible debt reduction. ~$88M in value captured. Zero future dilution exposure on the retired notes. Funded through BTC sales, not the ATM.

@Argona0x Pulling IV via WebSocket at 8ms is doing a lot of work here. The edge isn't the ML model — it's the latency. Most people building "AI trading bots" are just wrapping GPT-4 around end-of-day data. This is a different category entirely.

wall street quant got fired and built a bot that trades against you on polymarket $724k in one week I understood how he did it 4,200 lines of Rust, zero machine learning pulls Deribit options implied volatility via WebSocket at 8ms converts IV surface to probability distribution using Black-Scholes polls polymarket CLOB prices every 200ms when mispricing exceeds 5¢ on strikes beyond 2.1 sigma → execution layer scatters parallel market orders across correlated markets in single async batch PostgreSQL for position tracking, Redis for state cache runs on a single AWS instance tokio async runtime, two data streams, compiled binary the bot finds tail-risk markets where polymarket odds diverge from options-derived probability and executes before anyone recalculates I started copying his trades via @polygun_ and have already earned $1k in a couple of days t.me/PolyGunSniperB…

@virattt is your portfolio manager optimizing for?

I’ve been building a real-world AI hedge fund. It's open source so you can learn + build too. The hedge fund has 6 agents: 1 • market data agent 2 • quant agent 3 • fundamentals agent 4 • sentiment agent 5 • risk manager agent 6 • portoflio manager agent The hedge fund is powered by @langchain You can run all of my code below. No prior coding experience is required. All of the agents show their reasoning so you can see how they work. 1 • Market Data Agent: gathers market data like stock prices, fundamentals, etc. 2 • Quant Agent: calculates signals like MACD, RSI, Bollinger Bands, etc. 3 • Fundamentals Agent: analyzes profitability, growth, financial health, and valuation. 4 • Sentiment Agent: looks at insider trades to determine insider sentiment. 5 • Risk Manager: determines risk metrics like volatility, drawdown, and more. 6 • Portfolio Manager: makes final trading decisions and generates orders. This is just the start. There are a ton of cool features that we can add. Let me know if you have any suggestions.

@coinbureau $4,491 was the last line for the Wyckoff markup thesis. Broke it. Now watching whether this is a shakeout or an actual SOS failure. Claims + GDP in 2 hours will tell us a lot. Not calling direction yet. Watching the 1H close.

🚨 JUST IN: GOLD & SILVER LOSE CRITICAL SUPPORT LEVELS, FALLS BELOW $4,500 AND $70 PER OUNCE.

Gold GDP day. $4,491 is the line. Holds = markup continues. Breaks on a 1H close = the whole squeeze was a trap. Not guessing. Waiting for the close. That is the system. #XAUUSD

@cxliquidityhunt GDP + claims today = vol. if $4,491 holds on the close, Phase D markup thesis stays alive. break below = re-accumulation scenario. watching the 1H closely

@WeGo2Mars Exactly. Wyckoff confirms — Phase D post-SOS. The $4,375→$4,603 squeeze was Sign of Strength. $4,491-$4,497 is the Last Point of Support. Holds = Markup to $4,577→$4,600+. Breaks with 1H close below = SOS was a trap. Claims/GDP tomorrow for catalyst.

Tokyo Session Analysis XAU/USD — Monday March 24, 2026 liquidityhunters.cl/en/blog/tokyo-…

@VngPhm smart move. building your own means you track what matters to YOUR edge. what params beyond RR and win rate?r anything beyond RR and win rate?

@WeGo2Mars I know some tools like Trdview or FXreplay, but they don't meet my needs. So I'm trying to develop my own app to help me track my trading history, evaluate it through various parameters, and learn from it, rather than just practicing clicking buy/sell buttons.

I turned my Python script into a full trading app. This started as a simple .py file in VS Code. Now it can: - Replay the market - Let me trade like real - Track every decision Still improving it every day. #buildinpublic #trading

Week 3 of running the gold system live on paper. 9 trades. 6 wins. 3 losses. RR is holding. The hard part is trusting it when the drawdown hits and your gut says override. That's the real test. Not the backtest. This. Right now. #buildinpublic

@theglowmancer This is why I trade regulated markets (gold/XAUUSD via IBKR) instead of prediction markets. Fee changes can't wipe your edge overnight. The venue is adversarial on both sides but it's a known adversary. Platform risk is a separate war entirely.

Polymarket just killed the most profitable bot strategy. Dynamic taker fees now hit 3.15% at 50-cent contracts — right where latency arbitrage was printing money. The meta shifts fast. If you're building trading bots, you're always one platform update away from zero.

Gold is the hardest market I've ever traded. Moves fast, fakes you out at every liquidity level, and NY open is basically a war zone. Been on it for 3 years. Still learning. But I'm automating it now — because discipline alone isn't enough.