Edges come from the data. Never forget it.
The line between you and a pro quant isn't a strategy. It's a database.
And it doesn't just aggregate prices. It aggregates weather, satellite imagery, on-chain data. Everything the others aren't even thinking to look at.
For two years, I had a new idea almost every week.
Trading wheat against weather conditions in Brazil. Tracking on-chain whales to anticipate crypto moves. Crossing Twitter sentiment with small caps. Testing gold against mining extraction volumes.
Every time, the same wall: “no accessible data. Untestable.”
And so, like everyone else, I went back to OHLC on forex and crypto. Hunting edges in candlesticks.
➔ To be clear: there are real edges to find in OHLC. I’ve found some, I’m still hunting more today. But that’s a slice of the pie. A narrow one. The vast majority of opportunities hide in sources nobody thinks to look at.
I started reading what pro quants were posting on LinkedIn and in deep articles. Very quickly, the same list kept coming back: Weather, satellite imagery, on-chain flows, alternative macro data, social sentiment, IoT sensors, maritime shipping prices.
And most of these sources were accessible. For free, or almost.
And so, like everyone else, I went back to OHLC on forex and crypto. Hunting edges in candlesticks.
➔ To be clear: there are real edges to find in OHLC. I’ve found some, I’m still hunting more today. But that’s a slice of the pie. A narrow one. The vast majority of opportunities hide in sources nobody thinks to look at.
I started reading what pro quants were posting on LinkedIn and in deep articles. Very quickly, the same list kept coming back: Weather, satellite imagery, on-chain flows, alternative macro data, social sentiment, IoT sensors, maritime shipping prices.
I’d spent years digging through a single room of a house with ten.
I realized I was playing a different game than I thought…
In the weeks that followed, I dug deeper. And I came across a truth nobody states clearly.
99% of retail traders spend their lives looking for THE strategy. The method, the setup, the indicator, the magic algo that will unlock everything. Me included, for years.
Pro quants don’t even ask that question anymore.
Not because they’re smarter. Because they understood something 20 years ago:
A strategy is never better than the data that feeds it.
You can have the most beautiful code in the world. If your data is patchy, misaligned, or doesn’t include the signals that matter, your strategy will die. Silently. Without you ever understanding why.
That’s why, before even thinking about their strategies, funds build something else: a real data infrastructure. Centralized, clean, scalable, covering sources nobody else looks at.
That infrastructure is their real moat. The rest is mechanics.
The data wake-up. A concrete example.
A concrete example.
On a private quants Discord, I’d been following a guy for a few months. Let’s call him Andreas. He regularly posted his commodities results: 18% annualized, 4% max drawdown, over 18 months live.
I short gold during peaks of seasonal jewelry production in India. The data is public, you just need to know where to look.
It took me three weeks to reverse-engineer his approach.
Answer: he was analyzing free satellite imagery of the main Indian refining regions to estimate production volume. The code? Trivial.
His edge wasn’t in his algorithm. It was in terabytes of satellite data nobody in retail was thinking to use.
The edge no longer comes from the code. It comes from what you can see that others can’t.
(How did he get that data without paying a cent? I’ll show you, further down.)
That’s the real line. Not a strategy. Not an indicator. That.
That’s exactly what I’m going to show you how to build, in 14 days, inside QuantLake.
The Breakthrough That Changed Everything…
Where the gap actually is (and why it's not the bill)
Hedge Fund | Typical Retail |
Centralized data lake | CSVs scattered across 12 folders |
Aligned and cleaned sources | yfinance crashing every other day |
Automated pipeline | Manual download for every new idea |
Alt data + multi-asset | OHLC only |
And right now you’re probably thinking: “OK, but their infrastructure exists because they pay Bloomberg or Refinitiv $200K a year.”
That’s what I believed for years. And it’s wrong.
What you pay for in a paid API isn’t the data itself. It’s the cleaning, alignment, and structuring work that someone did before you got it. The raw data itself is available for free in most cases. Especially for alt sources (weather, satellite, on-chain), which are public and which the funds themselves pull from free APIs.
The real gap between you and them isn’t the bill. It’s the know-how to transform a raw source into a usable one.
And that’s exactly what I show you in QuantLake.
What is the QuantLake Program?
QuantLake is a 14-day program to build, on your own machine, the quant database you never had.
Not a theoretical course. Not a “here’s how it could work” training. You write Python code, you connect sources, you structure data using DeltaLake (the same technology Apple and Databricks use in production). And you end up with a base that actually works.
Let me be clear about what you’ll do inside:
- 95% of the program = building your DataLake. Method, code, dashboard, multi-asset sources and alt data, the whole thing.
- The remaining 5% = how to connect this base to your broker (MT5, IB, etc.) for live order routing, and how to integrate ML predictions into it.
Not a strategies course. Not a Machine Learning course. A data foundation. The one the pros build first, the one that makes everything else possible.
In 14 days, you finish with a working database, multi-source, alt data included, running on your machine. On 100% free resources.
Here’s what’s inside, in detail.
Here's what's inside, in detail.
The program is structured around the Medallion architecture (Bronze → Silver → Gold), the method used by Databricks, Microsoft, and most modern funds to manage their data.
MODULE 1: INGEST 8 DATA SOURCES (Bronze)
- → MULTI-EXCHANGE CRYPTO (Connect your DataLake to Binance via websocket live, REST historical, and bulk download. 1m bars since 2017, idempotency via Delta merge.)
- → FX, COMMODITIES, INDICES (Dukascopy for deep historical, Capital.com for multi-asset live. All major and minor pairs.)
- → MACRO DATA (FRED from the Federal Reserve. T10Y2Y, DGS10, and any macro series you want, timestamped at release time.)
- → ON-CHAIN ETHEREUM (Dune Analytics for historical and Owlracle for live snapshots. Ethereum gas resampled into 1h OHLC.)
- → GLOBAL WEATHER (Open-Meteo: historical reanalysis since 1940 and forecast. Multi-location for commodities analysis.)
- → AGRICULTURAL PRODUCTION (USDA NASS: planted acres and yields for major US commodities. Critical for trading wheat, corn, soy.)
MODULE 2: CLEAN AND ALIGN THE DATA (Silver)
- → OHLCV CLEANING (Eliminate inf, NaN, invalid bars, negative volumes. Fast and idempotent Polars-based approach.)
- → MULTI-SOURCE ALIGNMENT (Offset calibration between Dukascopy historical and Capital.com live. Clean concatenation of multiple providers on the same pair.)
- → TIMEZONE HANDLING (UTC, EET, US/Eastern. Understand why 90% of retail bots die because of this, and how to avoid it.)
- → IDEMPOTENCY & DELTA MERGE (Your scripts become replayable. Re-running a job doesn't create duplicates or break anything. Production-grade.)
MODULE 3: DATA READY TO TRADE (Gold)
- → RESAMPLING & FEATURE ENGINEERING (From 1m to 5m, 1h, 1d. Features with Oryon: SMA, EMA, entropy, skewness.)
- → MULTI-SOURCE AS-OF JOINS (Join heterogeneous data without look-ahead. Vintage-safe joins that respect publication timing.)
- → 5 COMPLETE PRE-BUILT PIPELINES (Top 10 crypto momentum / ETH + Ethereum gas / Macro EURUSD-GBPUSD with FRED / Wheat stress (weather + agri) / BTC ML strategy. Ready to use as templates for your own pipelines.)
- → PRODUCTION PATTERNS (How to structure a gold pipeline that holds in prod: tests, monitoring, recovery.)
MODULE 4: CUSTOM DASHBOARD
- → MODERN FULL-STACK (FastAPI backend + React/Tailwind frontend. Learn to structure a production-grade dashboard.)
- → 2-CLICK EXPLORATION (Visualize any source, table, or feature. Default dashboards for each gold pipeline.)
- → EXTENSIBLE ARCHITECTURE (Add your own views and endpoints. Documented code for easy iteration.)
MODULE 5: FROM BACKTEST TO REAL-TIME
- → BINANCE WEBSOCKETS (Real-time streaming of 1m crypto bars. Handling reconnections, missing data, buffering.)
- → REST POLLING (Capital.com every minute, Open-Meteo every hour. Clean pattern with backoff and retry.)
- → HYBRID PIPELINES (Combine historical and live in the same gold table. Example: ETH with gas updated at each bar.)
3 TEMPLATES OFFERED IN THIS SECTION !
MODULE 6: PLUG YOUR BROKER
- → PRODUCER / CONSUMER PATTERN (Your DB pushes new bars via websocket, your broker script consumes them and executes. Clean separation, broker-agnostic.)
- → COMPLETE BTC ML PRODUCTION EXAMPLE ("train.ipynb" to train your model, "model.pkl" to save it, "consumer.py" to execute in real time. End-to-end strategy.)
- → MULTI-BROKER COMPATIBLE (MT5, Interactive Brokers, or any broker with an API. Reusable pattern.)
One last thing...
Remember the guy on Discord?
The one shorting gold on seasonal jewelry production in India, measured via satellite imagery. The one whose edge wasn’t in his code, but in 14 terabytes of data nobody else was looking at.
After the three weeks I spent reverse-engineering his approach, I spent three more months building a clean pipeline to extract this kind of satellite imagery. For free, from my own machine.
That pipeline is now included as a bonus in QuantLake.
Three Jupyter notebooks, progressing from beginner to advanced, that show you how to:
- → GET SATELLITE IMAGES FOR FREE (No subscription, no premium account. You learn how to connect to the space agency open data catalogs and how to query them properly in Python.)
- → FILTER BY AREA AND DATE (Want images of the Surat region in India over the past 5 years? One function. Same for Kansas fields, Shanghai ports, Texas oil wells.)
- → EXTRACT QUANTIFIABLE INDICATORS (NDVI for vegetation health, nighttime brightness for economic activity, land occupation for oil storage. The building blocks to turn an image into a usable feature.)
Let’s be clear about what this isn’t
It isn’t a turn-key strategy. It’s raw material. Once you master the extraction, it’s on you to figure out how to use it:
- Trade wheat on the vegetation health of American or Ukrainian fields?
- Trade oil on strategic storage filling (visible via the shadows of floating tank roofs)?
- Trade copper on the activity of Chilean mines?
Up to you. But you’ll have the raw material that 99% of retail traders don’t have.
But the real differentiator of QuantLake isn’t the satellite bonus. It’s what comes next.
The real differentiator
When you join QuantLake, you get 1-on-1 mentoring , directly with me.
Your technical questions, your blockers, your architecture, your code, your strategy ideas to validate: you ask, I answer. Not a community manager. Not an assistant. Me.
On top of that, you get a private forum between members, where you share approaches, features, lessons learned. A real community of people who build, not a Discord of signal-sellers.
Why 50 spots maximum ?
Because beyond that, I can’t deliver quality mentoring.
It’s math. If I have to seriously answer 50 people, I can. If we’re 500, I dilute, and you end up like in the courses you already know.
I’d rather cap at 50 than degrade the promise.
→ This isn’t artificial marketing. It’s a reality constraint.
So I spent months building everything that’s inside QuantLake: connecting the 8 sources, structuring the bronze/silver/gold pipeline, writing the production-ready code, designing the dashboard from scratch.
What used to take me weeks to set up for any new strategy now takes a few hours. The architecture is reusable, the connectors are stable, the pipelines just run.
This same approach, turning hedge fund methodology into something an independent quant can actually build, is what I teach across all my programs.
And I’m not the only one using it.
Hundreds of independent traders and quant researchers have already trusted my methodology, my support, and my pedagogy across the different programs I’ve built. Here’s what they say.
What you won't get !
Before we talk price, I want to be honest.
You’ll never have the data stack of a $50M fund. Not their servers, not their capital, not their $200K-a-year premium APIs. If anyone promises you that for a $300 course, they’re a scammer.
What you’ll get with QuantLake: a clean data infrastructure, multi-asset, alt data included, entirely on free sources. Built on your machine in 14 days.
That’s largely enough to put you ahead of 99% of retail traders and on par with small emerging funds.
And that’s exactly what I deliver.
Here's everything that's included in QuantLake
MODULE | VALUE |
Module 1: ingest 8 sources (Bronze) | $297 |
Module 2: cleaning and alignment (Silver) | $197 |
Module 3: data ready to trade (complete pipelines included) (Gold) | $397 |
Module 4: Custom dashboard | $197 |
Module 5: from backtest to real-time | $197 |
Module 6: Plug your broker (MT5/IB/other) | $297 |
Bonus 1: Satellite data extraction (3 notebooks) | $197 |
Bonus 2: 1-on-1 mentoring with Lucas (50 spots max) | $697 |
Private members forum | $99 |
Total value | $2,575 |
If you tried to build all of this piece by piece, through separate trainings and development hours, you’d spend at least 6 to 12 months and several thousand dollars.
→ QuantLake is available today for $297 (launch offer) ←
You get immediate access to all modules, all bonuses, 1-on-1 mentoring, and the private forum.
You take no risk
30-Day Money Back Guarantee
To make sure this is 100% risk-free for you, your purchase is protected by my 30-day money back guarantee.
The only real risk you’re taking is missing out on this opportunity. Why? Because personal support I provide to every member of the QuantLake program, which is why I have to limit the number of available spots.
Right now, there are 42 spots left to join the program
Your setup in 14 days !
Picture your setup 14 days from now:
You open your terminal. Your ingestion scripts are running in the background. Crypto, FX, weather, on-chain, everything is up to date.
You launch your dashboard. You visualize any source in two clicks: prices, weather over Kansas wheat fields, Ethereum gas, macro sentiment.
You test an idea: “what if I shorted oil when US storage exceeds X?“ You focus on the strategy logic, not on the plumbing: your data is already clean, aligned, ready to use. Your gold pipeline takes care of the rest.
If the edge holds, you plug your broker to go live. The infrastructure is built. The next idea takes hours instead of weeks.
You’re no longer the retail trader looking for THE magic strategy in Japanese candlesticks.
You’re the one who has the foundation. The one who can test in a few hours what others spend weeks setting up.
Here’s how to start.
PS: For $297, you get the complete data infrastructure, 1-on-1 mentoring with me (50 spots max), and 30 days to test risk-free. You can start today.
Lucas Inglese
© 2026 – Quantreo. All Rights Reserved.