azoroso

🎯 An agentic AI that writes its own factor signals -> and backtests them? A paper that came across my radar at Noax this week: Huang and Fan build a fully autonomous framework for systematic factor investing using agentic AI. No manual prompting at each step: the system formulates signals, validates them out-of-sample, and applies economic-rationality filters on its own. The results it reports: annualized Sharpe of 3.11 and returns of 59.53% on US equities. My read: those numbers are almost certainly not replicable in a live setting. The gap between backtest Sharpe and real-world Sharpe is especially wide when the system itself is generating the signals. The risk of overfitting is enormous. But the architecture is what's interesting. The idea of closing the loop, signal generation, validation, and refinement in a single autonomous pipeline, is where systematic investing is likely heading. Worth reading as a directional piece, not as a performance claim. 📄 Paper: arxiv.org/abs/2603.14288 --- → Join the newsletter: ivanblanco.ai/newsletter ---

I backtested a simple system on Bitcoin since 2023. 100% or 0% allocation. No leverage. No hedging. +965% return −19% max drawdown In market 51% of the time Buy & hold did +322% Datasets included bitcoin, altcoins, and stocks.

I never thought the Money Flow Index would be any good. But when I tested 100 indicators on 3,500 strategies, the truth came out. And it really stood out. It ranked #1 in the volume cluster and #2 overall. It consistently improved net profit, drawdowns, and win % across 15 years of data. I was pretty impressed! MFI tracks smart money really well - at least based on my thorough testing. Do you use MFI yourself? Watch the video and get the free resource in the comments.

I’ve been trading for ~30 years. First half: fully discretionary, living inside futures microstructure. It worked—until algos started exploiting the same patterns and reacting in microseconds. Edge decay was real. So ~10 years ago I switched to systematic. Now I run many uncorrelated strategies in parallel without babysitting screens all day. I wouldn’t go back. My biggest unlock: reusability of know‐how. When I finish a new system, I plug it into a ready workflow in minutes. It monitors itself; I move my brain to the next big thing. Here’s the playbook I wish I had from day one: - Framework (design once → reuse forever) - Data → clean, feature, label. - Hypothesis → simple, testable edges (breakouts, momentum, mean reversion). - Validation → IS/OOS, realistic costs/slippage. - Risk → position sizing, max heat, portfolio exposure caps. - Deploy → automated orders, fail‐safes. - Monitor → health dashboards, kill‐switch rules, mobile app. - Iterate → new systems slot into the same pipeline. Principles that compound: - Many small, independent edges > one “genius” setup. - Process beats prediction. - Shipping beats perfecting. Discretionary taught me markets. Systems gave me scale.

The $100K Rule 1. Remove the TradingView Default Volume 2. Search for "VolUSD" in the indicator library and add the indicator to your chart. 3. Click the settings icon (gear icon) on the VolUSD indicator and match the image. Only trade coins with at least $100,000 average VolUSD per 1-minute candle on Binance.

Factor investing is what made me stand out at JPMorgan. But it took me years to master the information coefficient. In 1 minute, I'll teach you the 10 things you need to know (that took me 1 year to learn). Let's go:

How did I manage to maintain an 85% win rate during this drop, anticipate the bear market three months ago and exit my spot positions, taking only one loss in the process and six wins, all live? Here are the strategies I use to avoid missing the next bottom and to stop longing every higher low on $BTC

Based on The Man Who Solved the Market, I decided to add p-value to determine the probability that a strategy's performance is just noise. Jim Simons was using p-value to assess whether an edge was real or random, So I'll add that to my trading journal as well. It's very hard to determine when to remove a strategy from the portfolio. Some people use rolling Sharpe for that. Now, who knows if p-value still fully holds up They were using it in the '90s But Black Scholes Merton is old too, and quants are still using it.

Things You'll Regret Not Doing as a Trader: • Learning position sizing fundamentals properly • Backtesting a lot of ideas • Understanding risk management early (drawdown math, correlation,etc) • Learning market microstructure basics (slippage, spread, liquidity, order types) • Practicing proper validation (walk-forward, Monte Carlo, parameter sensitivity) • Deploying strategies consistently, not just collecting backtests • Taking health seriously (sleep, exercise, stress management) because burnout will bill you later • Staying close to 2-5 high quality traders • Taking calculated risks: automating execution, switching brokers, going full-time, etc • Learning to embrace boredom so you don't need constant dopamine

Article Summary: How to fix your entire life in 1 day

You can now turn Claude Code into a full development stack. A new update ships a CLI that installs agents, commands, and integrations in minutes, with 14.8k GitHub stars. 𝗖𝗹𝗮𝘂𝗱𝗲 𝗖𝗼𝗱𝗲 𝗰𝗮𝗻 𝗻𝗼𝘄 𝗯𝗲 𝗰𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗲𝗱 𝗹𝗶𝗸𝗲 𝗮 𝗿𝗲𝗮𝗹 𝘁𝗲𝗮𝗺 It packages roles, workflows, and tools as reusable components. You install them with one command. No manual prompt wiring or copy pasting. • Agents for code review, security, and performance • Slash commands for testing, optimization, and checks • Hooks for pre commit and post task automation 𝗜𝘁 𝗮𝗱𝗱𝘀 𝗼𝗯𝘀𝗲𝗿𝘃𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝘁𝗼 𝗔𝗜 𝗰𝗼𝗱𝗶𝗻𝗴 It tracks sessions, state, and usage in real time. You can inspect live conversations locally or remotely. • Analytics with performance metrics • Health checks for misconfiguration • Plugin dashboard with permission control 𝗜𝘁 𝘀𝗼𝗹𝘃𝗲𝘀 𝗿𝗲𝗽𝗲𝗮𝘁𝗮𝗯𝗶𝗹𝗶𝘁𝘆 𝗮𝗻𝗱 𝘀𝗰𝗮𝗹𝗲 Instead of one off prompts, you install standardized behavior. That lets teams share setups, debug faster, and onboard consistently.

People keep asking me for advice on how to best use Claude Code. Just read this. Couldn't have said it better myself.

5/ With multi-chart support, you can watch how correlated assets evolve in real time, side by side. In this view, BTC, MSTR/USD, and multiple MicroStrategy-related prediction markets are displayed together, despite belonging to different asset classes. They’re shown in one unified workspace because they share the same underlying risk and narrative drivers. When price moves in one market, you can immediately see how probabilities adjust elsewhere. This makes cross-asset relationships explicit instead of inferred, and lets you trade narratives as they propagate across markets.

We’re excited to launch a completely new way to see prediction markets. Today we’re rolling out a complete trading stack for prediction markets, from visualization to execution. This unlocks a fundamentally different way to see how money positions before odds move. Here’s what that looks like in real time: a large sell wall appears in the Portugal Presidential Election market hours before the odds moved down sharply.

hey @grok Act as a professional, systematic trader. Design a fully rule-based trading model to trade Gold (XAUUSD). The model must clearly define: • Market conditions required (trend, range, volatility, news filters) • Best trading session (Asia, London, New York) • Exact setup logic (what must happen before a trade is allowed) • Precise entry rules (no discretion, candle/level/confirmation rules) • Stop-loss placement (fixed or dynamic, and why) • Take-profit logic (targets, partials, or trailing if any) • Risk-to-reward framework (minimum R:R and how it’s enforced) • Position sizing rules (risk per trade) • Trade management (when to hold, when to exit early, when to do nothing) • Invalidation rules (when the setup is no longer valid) Explain it step by step, as if the goal were consistency and capital preservation, not overtrading. No motivation. No general advice. Just the actual model and logic.

Gold surprised me. A swing breakout on GC has been one of the cleanest performers I’ve tested - and the rules are embarrassingly simple: - Swing = 1x ATR move - Long on swing high break - Short on swing low break - Exit on 5th 60-min bar (3rd/4th similar) - Fees: $20 RT per contract 21.7y results: 1,454 trades - 54.6% win - PF 1.48 - Sharpe 1.04 - Max DD 11.4%

One of the biggest mistakes I see in systematic trading: Using fixed stops instead of ATR-based stops. Fixed stops are easiest to overfit, don't account for volatility, don't make logical sense for most strategies.

A Lite Thread about Central Bank and the Forex Market.