realtest ai trading strategy mhptrading

Full Course

Build RealTest Strategies with Claude Code AI

Save hundreds of hours of manual RealTest work by learning the AI workflow for coding, testing, optimising, and validating strategies.

Enroll Now

No Manual Coding Required

Lifetime Access + Updates

Certificate Of Completion

realtest claude code ai trading course tutorial
anthropic ai trading realtest
claude ai trading realtest
notion ai trading realtest
ai trading realtest logo png
Gradient background transitioning from light gray to orange
realtest ai claude code trading backtest

AI Is Changing Algorithmic Trading

Claude Code can write scripts, run tests, read results, and document your research.

This course shows how to use it with RealTest so you can keep pace.

What You’ll Learn

Set up Claude Code with RealTest

Turn trading ideas into clear strategy plans

Generate RealTest scripts with AI

Fix script errors automatically

Run backtests with AI

Analyse backtest results with AI

Optimise parameters without curve fitting

Run walk-forward validation with AI

Combine strategies into portfolios

Create automatic Notion reports

Use ready-made prompt templates

Build a repeatable AI research workflow

realtest ai trading

Work Smarter, Not Harder

Flexible online training designed to help you supercharge your trading strategies with AI.

Review Backtests From Your Phone

Start the AI workflow once.

Claude Code runs RealTest, reads the output, writes the analysis, and saves the report to Notion while you are away from the desk.

When it is finished, you can open the result on your phone.

ai realtest trading backtest

What’s Included

8 Core Modules + Reference Sections

7 Walkthrough Videos

Downloadable Workflow Files

8 Copy-Paste AI Prompts

Course Modules

  • Install Claude Code for the course workflow.
  • Adding the SetupAlpha course files
  • Creating the strategies folder for AI-generated backtests
  • Connecting Claude Code to RealTest
  • Verifying that Claude can see the files, documentation, and output setup
  • Use AI to plan a strategy before writing code.
  • Turn a rough idea into a complete strategy plan.
  • Define exactly what market behavior the strategy is trying to capture.
  • Run a planning session from idea to finished plan.
  • Check for missing or vague assumptions before backtesting.
  • Identify the 7 steps from finished plan to result files.
  • Understand what the AI does automatically in the background.
  • Locate the failed step when a run produces an error.
  • Add permanent rules for future AI behaviour.
  • Adjust the AI command allowlist safely.
  • See how common RealTest errors are handled.
  • Identify the 3 result files RealTest produces.
  • Test whether profits come from a real edge or only a few outlier trades.
  • Read the most important performance numbers together.
  • Use the 3 analysis prompts after a backtest.
  • Use a pass/fail checklist to decide whether the strategy is worth optimizing.
  • Recognise when an analysis answer is too vague.
  • Use AI to plan and run parameter optimisation.
  • Learn why the goal is a stable performance range, not the single best result.
  • Choose between the 4 optimisation modes.
  • Count parameter combinations before running a test.
  • Separate edge-defining parameters from cosmetic ones.
  • Spot the 5 warning signs of overfitting.
  • Choose the right score column for the strategy.
  • Configure walk-forward optimisation.
  • Understand what walk-forward testing measures.
  • Run the 2-step walk-forward process.
  • Read in-sample and out-of-sample result rows.
  • Interpret the gap between IS and OOS performance.
  • Know when sample size makes WFO less reliable.
  • Decide whether the strategy is robust enough or too overfit to trust.
  • Use AI to evaluate whether strategies should be combined.
  • Understand when combining actually reduces risk.
  • Run the 4 diversification checks.
  • Choose between the 4 RealTest combination methods.
  • Verify the capital math before finalising a portfolio.
  • Walk-forward test the capital split between strategies.
  • Connect Notion to Claude Code.
  • Generate a structured backtest report with one prompt.
  • Build reports that include the 6 standard sections.
  • Save strategy logic, numbers, analysis, and code together.
  • Create a searchable record of your backtest research.
  • Understand where AI is useful in the strategy research process.
  • Learn why AI can help structure, code, test, and analyze ideas.
  • Separate trading logic, backtesting workflow, and market judgment into the right responsibilities.
  • Avoid treating AI output as proof that a strategy is profitable.
  • Learn the limits of AI when working with financial markets, live trading, and real capital.
  • Use ready-made prompts for each part of the strategy research workflow.
  • Start new strategy ideas, backtests, optimizations, validations, and reports faster.
  • Learn what details to include so the AI gives useful, specific output.
  • Avoid vague prompts that lead to incomplete strategies or unreliable analysis.
  • Reuse the templates as a starting point and adjust them for your own trading ideas.
  • Learn how to keep developing new strategy ideas after finishing the course.
  • Build a habit of testing, rejecting, improving, and documenting strategies.
  • Use your own market observations as the starting point for future research.
  • Continue expanding the process into more markets, strategy types, and portfolio combinations.
  • Download the SetupAlpha course files used throughout the workflow.
  • Includes the AI instruction files, RealTest references, error log, and output template.
  • Provides the folder structure needed for strategy creation, testing, optimization, and reporting.
  • These files are required for the course workflow to run correctly.
  • Zip files (257 KB).
  • Quick definitions for terms used in the course.
  • Use it as a reference when a module introduces an unfamiliar term.

Course Feedback From Traders

claude code ai realtest course

I’ve just completed SetupAlpha’s Claude course, and it is absolutely fantastic. I had tried to connect RealTest to Claude before on my own but really struggled. SetupAlpha’s clear and concise instructions made the integration work perfectly, and the custom .md files they created make everything just work seamlessly.
Claude's precision is vastly improved from a few months ago. At one point when I thought Claude had made an error, it actually proved that the AI was correct and I was wrong. I’ve already used what I learned to go through one of my portfolios that I was always a little concerned about regarding robustness, and it is looking much healthier now.

The biggest game-changer is that I can actually discuss strategies with Claude now and it truly feels like a real trading partner. I don’t have to worry about it making continuous stupid errors like it did just a few months ago, those are completely gone now, and it doesn’t just agree with whatever you say, it pushes back with valid arguments. I’m certain that SetupAlpha’s specialised .md files have gone a long way to making that level of accuracy happen.

I highly recommend this course to any trader looking to go further with AI and improve their strategies as well as develop new ones.

- Julian

I’m not a programmer, and RealTest has always been one of those tools I knew was powerful but could easily get stuck in.

What helped me most was the step-by-step setup.

Now I can take an idea, ask better questions, get a script started, and understand the output without feeling like I’m fighting the software/AI the whole time. That alone made the course worth it for me.

- Robin

The biggest thing for me was not getting stuck as much.

Before the course, I’d sit down with an idea and somehow end up spending most of the time on annoying little stuff, RealTest errors, bad prompts, checking the wrong file, or trying to figure out what Claude actually needed from me.

Now it’s more clear what to do next. I know how to set things up, what to give Claude, and how to check whether the output makes sense.

It saves a lot of wasted time. I can get back to testing the actual idea much faster.

- Didan

Who This Course Is For

  • RealTest users who want to add AI to their research workflow
  • Traders with strategy ideas they want to test faster
  • Systematic traders who want to spend less time writing scripts by hand
  • Traders who have tried AI for trading research and seen it produce unreliable results
  • Traders tired of generic AI answers that look confident but break in RealTest
  • Traders who want to save time on coding, debugging, reporting, and repeat testing
  • Traders who want a repeatable process for testing, optimisation, validation, and reporting
realtest software backtesting course tutorial

Requirements

realtest software mhptrading logo svg png icon setupalpha

RealTest

  • The course is built around RealTest and uses its folders, scripts, backtests, optimisation, and output files.

Claude Pro subscription

  • Required for Claude Code, which writes scripts, runs RealTest, reads results, and analyses the workflow.

Free Notion account

  • Used for the automatic backtest report module. The free plan is enough.

The AI Revolution in Trading

See how the industry's top minds view the impact of AI on productivity and strategy.

Many people talk about AI as a threat, but I've always seen it as a partner to help me in my decision making.

Ray Dalio

X Tweet

Generative AI is the next new thing in machine learning. This will be another wake up call for corporate America to focus on how to create productivity gains.

Ken Griffin

Bloomberg Interview

My CTO comes to me and says I can save the firm $25 million by using these AIs to improve our efficiency.

Steve Cohen

CNBC Interview

AI's coming for me now. + It's annoyingly better than me.

Gliff Asness

Bloomberg Interview

AI will undoubtedly boost productivity for everyone. It’s like each of us getting a team of highly capable, tireless assistants that can process a vast amount of information, synthesize insights, and automate mundane tasks.

Jin Choi

Two Sigma Blog

AI is a long-term secular trend, don't see it reversing.

David Solomon

CNBC Interview

Many people talk about AI as a threat, but I've always seen it as a partner to help me in my decision making.

Ray Dalio

X Tweet

Generative AI is the next new thing in machine learning. This will be another wake up call for corporate America to focus on how to create productivity gains.

Ken Griffin

Bloomberg Interview

My CTO comes to me and says I can save the firm $25 million by using these AIs to improve our efficiency.

Steve Cohen

CNBC Interview

AI's coming for me now. + It's annoyingly better than me.

Gliff Asness

Bloomberg Interview

AI will undoubtedly boost productivity for everyone. It’s like each of us getting a team of highly capable, tireless assistants that can process a vast amount of information, synthesize insights, and automate mundane tasks.

Jin Choi

Two Sigma Blog

AI is a long-term secular trend, don't see it reversing.

David Solomon

CNBC Interview

claude code ai realtest course

10x Your Backtesting

AI won’t replace you. Someone who knows how to use AI will.

$1559 $790
Enroll Now
One-time payment • Lifetime access

Frequently Asked Questions

What is the RealTest and Claude Code AI course?

This course teaches RealTest users how to use Claude Code to plan trading strategies, generate RealTest scripts, run backtests, analyze results, and document research in a repeatable workflow.

Who is this course for?

It is for systematic traders, RealTest users, and strategy researchers who want to use AI to speed up strategy development, testing, optimization, and reporting.

Do I need RealTest before taking this course?

Yes. The course is built around RealTest workflows. You should have RealTest installed or be ready to use it while following the course.

Do I need to know how to code?

No. Claude Code writes the RealTest scripts for you. You do not need to manually write RealTest code from scratch, but you should still review the logic and results.

Do I need Visual Studio Code, Cursor, Antigravity, GitHub, or another coding editor?

No. The course uses the Claude Code desktop app connected directly to RealTest. You do not need VS Code, Cursor, Antigravity, GitHub, or a separate coding environment.

Can I use OpenAI Codex, Gemini, Cursor, or another AI tool instead of Claude Code?

Yes, the ideas transfer, but the course is demonstrated with Claude Code. Other AI tools may handle files, permissions, commands, and workflows differently.

What does Claude Code do in the RealTest workflow?

Claude Code can help plan strategies, write RealTest scripts, run tests, read output files, analyze results, find errors, and prepare structured research reports.

What will I learn to build with RealTest and AI?

You will learn how to turn trading ideas into testable RealTest strategies, run backtests, analyze results, optimize parameters, validate robustness, and document findings.

Does the course teach strategy planning before writing code?

Yes. The workflow starts by defining the trading idea, market behavior, rules, assumptions, and validation checks before generating a RealTest script.

Does the course teach backtest analysis and validation?

Yes. You learn how to read RealTest result files, check whether performance comes from a real edge or outlier trades, and decide whether a strategy is worth improving.

What files and templates are included with the course?

The course includes AI instruction files, RealTest references, an error log, output templates, and a folder structure for testing, optimization, validation, and reporting.

Does this course provide trading signals or guarantee profitable strategies?

No. The course teaches an AI-assisted research and backtesting workflow. It does not provide investment advice, trading signals, or guaranteed profitable strategies.