Python trading · in development

Trading automation.
Built with evidence.

A Python-powered automated trading system in development, with licensed access planned for independent traders.

Explore strategies through historical testing, order-flow analysis and configurable risk controls. Review the evidence behind each decision before considering live deployment.

Built on Python & NautilusTrader.
Customer release remains under validation.

Python / NautilusTraderDevelopment code
Conceptual trading-system architectureResearch code connects market data, strategy analysis, risk checks and historical test records. Two planned customer delivery options branch below: an executable or a server-to-client bridge. Live operation is pending validation. This is a conceptual diagram, not a product screenshot. Market dataStrategy analysisRisk checksHistorical test records InputsLogicRiskEvidence
Executable Planned deliveryClient bridge Planned delivery
Conceptual architecture, not a product screenshot.
Customer delivery is planned; live operation is pending validation.

Understand the strategy.
Inspect the risk.

Connect trading ideas to historical tests,
risk checks and reviewable records.

From a trading idea
to evidence you can review.

The development code connects strategy research, historical simulation and risk checks. Customer delivery and live operation remain pending validation.

  1. Study the market

    Explore footprint and order-flow inputs, including imbalance and absorption analysis, within the strategy research workflow.

  2. Test the strategy

    Run historical simulations with NautilusTrader. Inspect strategy behavior and execution assumptions before interpreting a result.

  3. Inspect the risk

    Review configurable position sizing, drawdown checks and trading-time constraints alongside test records and rejection reasons.

The assumptions
matter as much
as the result.

A useful test record identifies the strategy, data and costs. Missing assumptions make a result harder to interpret.

Switch between two synthetic examples to see how the record flags missing information.

Explore the validation approach
Illustrative interaction only.
No live orders or real performance data.
Robot testing workflow

Inspect the record
behind a test.

Illustrative preview · synthetic data

Choose an example record
Experiment RX-024Test record only
Research record traceA strategy definition connects to a dataset, execution assumptions and a review record. This diagram does not show trading performance.StrategyDatasetCostsReview
Strategy version
research-example-v1
Dataset
synthetic-quotes-v1
Cost assumptions
Spread and fee recorded

Record complete

0 missing inputs

This example records the strategy version, dataset and cost assumptions. The record can be inspected; the research still needs independent validation.

Inspect the evidence trace
  • Strategy version: research-example-v1
  • Dataset: synthetic-quotes-v1
  • Spread assumption: 2 bps (synthetic)
  • Fee assumption: 0.5 per unit (synthetic)

An illustrative test-record interaction, not a running trading system, live trades or actual backtest results.

A trading system.
A clear path to licensed access.

In development

Python-powered automation

An automated trading codebase built on Python and NautilusTrader, combining historical testing, footprint analysis and configurable risk checks.

Live operation and a customer release still require validation. Production access is not available today.

Planned

Licensing & early access

The intended offer is licensed access to the Python trading system, with documented setup and configuration guidance. Customer delivery is being defined.

Availability, supported setups, licence terms and price will be agreed before any purchase. No checkout or paid access is offered here.

A product you can understand.

Designed for independent traders who want an inspectable automation workflow. The intended offer is licensed access to the Python system, with setup and configuration guidance.

Product & delivery

Explore the analysis modules, decision workflow and planned executable or client-bridge delivery models.

Read the product overview ↗

Make test evidence
easier to understand.

The intended Claude API layer will explain bounded test reports, connect statements to source evidence and flag missing assumptions for human review.

See the planned Claude integration
  1. Report inputs

    Test summaries, settings and rejection reasons.

  2. Evidence explanation

    Source-referenced summaries and gaps to investigate.

  3. Human review

    Check generated explanations against the source.

Planned feature. Claude is not integrated into the current prototype.

Franco Mascarelo Ortiz

Founder & developer

I’m building a Python-powered automated trading system for future licensing clients. The work combines algorithmic strategy development, historical simulation, order-flow research and reviewable risk checks.

Mascarelo AI is currently a founder-led product project, before company incorporation. Development began in December 2025. The robot is being developed for future clients.

The current research baseline uses gold futures. Supported customer markets and integrations will be confirmed through validation.

Contact Franco

Before you get in touch.

Can I license the robot today?

It is a development prototype; paid or live access is not available today. Email Franco to discuss future licensing and early access. Availability, supported setups, deliverables and price would be agreed before any purchase.

Does the preview run the Python system or show actual results?

No. Every value in the preview is synthetic. It supports the explanation of a testing workflow: a record can reveal missing assumptions. It does not run the Python trading system, execute live trades or show actual backtest results.

How will Claude be used?

The planned use of the official Claude API is to help explain test reports and identify gaps, with source references and human review. Claude is not integrated into the current prototype. No Anthropic affiliation or endorsement is claimed.

How would licensed access be delivered?

Delivery options under evaluation include a local executable, or a client-side bridge that receives trading instructions from a server. These are planned options, not available customer services. Supported platforms, connection security, execution permissions and licence terms must be defined and validated before release.

Does Mascarelo AI provide investment advice?

This project develops a Python-powered trading system. The site and illustrative preview provide no personalized investment advice, trade recommendations or performance guarantees.

Explore future access
to the Python system.

Tell Franco about your trading workflow and the access you need. Discuss the delivery model, supported markets and licence terms before any purchase.

Discuss licensing & early access

Opens your email app.
No signup is submitted on this page.

franco@mascareloai.uk