# ENSEMBLE

> The asset allocation platform for the agentic era. Build any strategy, test it, and read exactly how it was built. Each build returns generated code, a sandboxed backtest, and a public tear sheet. Models hold assets, other models, or both. The generated code is yours to download; models rerun nightly; current allocations export in Altruist and Interactive Brokers formats. Backtests are hypothetical; ENSEMBLE is research software, not investment advice.

## For AI agents

- [MCP server](https://api.ensemble.markets/mcp): remote Streamable HTTP, OAuth 2.1 or an `ensemble_` API key as bearer. 52 tools generated from the REST API: models, simulations and analyses (`create_`, `list_`, `get_`, `update_`, `delete_`, `rebuild_`, `run_`, `list_*_runs`), runs (`get_run`, `compare_runs`, `promote_run`), builds, assets, backfills, optimizers, plus `submit_feedback`, `list_feedback` and `get_feedback`.
- [REST API](https://api.ensemble.markets/v1): API key or signed-in session as bearer; lists are `{data, total}`, variants of a read are `include=`, one error shape. [OpenAPI document](https://api.ensemble.markets/v1/openapi.json), [reference](https://api.ensemble.markets/docs).
- [CLI](https://www.ensemble.markets/cli): `npx ensemble.markets <noun> <verb>`, generated from the same OpenAPI document; JSON on stdout, `ensemble schema` lists every command.
- [Pricing as markdown](https://www.ensemble.markets/pricing.md): plans, rate card, and included usage.

## Product

- [Pricing](https://www.ensemble.markets/pricing): usage-based, $100 in credit to start, spending cap you control.
- [MCP server setup](https://www.ensemble.markets/mcp): connect Claude, Cursor, VS Code, and any MCP client.
- [REST API](https://www.ensemble.markets/api): create, backtest, and re-run models from your own systems.
- [Methodology](https://www.ensemble.markets/methodology): data, sandboxed execution, cost assumptions, deterministic re-runs.

## Systematic strategies, explained and backtested

- [Strategies](https://www.ensemble.markets/strategies): Plain-language explanations of the classic systematic strategies, each with its rules, its failure modes, a hypothetical backtest on ENSEMBLE, and the prompt that builds it.
- [Dual momentum: the rules, the evidence, and a backtest](https://www.ensemble.markets/strategies/dual-momentum): Dual momentum combines relative momentum (hold the recent winner) with absolute momentum (move to cash when the winner is falling). Here are the exact rules, why it works, when it fails, and a hypothetical backtest you can rerun.
- [Risk parity: how it works, how to size it, and a backtest](https://www.ensemble.markets/strategies/risk-parity): Risk parity weights each asset so that it contributes equal risk to the portfolio, instead of equal dollars. Here is the logic, the simple inverse-volatility version you can implement yourself, where it breaks, and a hypothetical backtest to rerun.
- [The 60/40 portfolio: rules, variants, and a backtest](https://www.ensemble.markets/strategies/60-40-portfolio): The 60/40 portfolio holds 60 percent stocks and 40 percent bonds and rebalances back to those weights. Here is why it became the default, how rebalancing frequency and bond duration change it, the permanent and all-weather variants, and a hypothetical backtest to rerun.
- [The 200-day moving average strategy: rules, evidence, and a backtest](https://www.ensemble.markets/strategies/200-day-moving-average): The 200-day moving average rule holds an asset when its price is above its long-term average and moves to cash when it is below. Here are the exact rules, the research behind them, the whipsaw problem, and a prompt that backtests it in under a minute.

## Backtesting, from first principles

- [Backtesting](https://www.ensemble.markets/backtesting): How to backtest a portfolio without code, what the metrics on a tear sheet mean, and the mistakes that make a backtest lie to you.
- [How to backtest a portfolio without writing code](https://www.ensemble.markets/backtesting/how-to-backtest-a-portfolio): A step-by-step walkthrough of backtesting a portfolio from a plain-language description: choosing the universe, stating the rule, running the test, reading the tear sheet, and deciding what to change. Takes about ten minutes.
- [Backtest metrics explained: CAGR, volatility, Sharpe ratio, and max drawdown](https://www.ensemble.markets/backtesting/metrics-explained): What each number on a backtest tear sheet means, how it is computed, what a reasonable range looks like for a diversified portfolio, and how the metrics mislead when read alone. Definitions for CAGR, volatility, Sharpe ratio, maximum drawdown, turnover, and time in market.
- [Common backtesting mistakes and how to avoid them](https://www.ensemble.markets/backtesting/common-mistakes): The ways a backtest lies: lookahead bias, survivorship bias, overfitting, ignored costs, short or regime-limited samples, and misread metrics. What each one looks like, how to detect it, and what a backtesting tool can and cannot protect you from.

## AI agents for investment research

- [AI agents](https://www.ensemble.markets/agents): How Claude, Cursor, ChatGPT, and custom agents build and backtest systematic strategies through a finance MCP server or the REST API.
- [What is a finance MCP server, and is there one for backtesting?](https://www.ensemble.markets/agents/finance-mcp-server): A finance MCP server gives an AI assistant tools for investment research. This guide explains how MCP works, what separates a data-lookup server from a backtesting server, hosted versus local servers, authentication, and the tools ENSEMBLE's server exposes.
- [How to backtest a strategy with Claude](https://www.ensemble.markets/agents/backtest-with-claude): Connect Claude Desktop or Claude Code to ENSEMBLE's MCP server and backtest a strategy in conversation. Setup for both clients, a real session from idea to tear sheet to revision, and the prompts that get good results from an agent.

## Backtesting tools compared

- [Compare](https://www.ensemble.markets/compare): Factual comparisons of portfolio backtesting tools: what each one does well, where it stops, and when ENSEMBLE is or is not the better fit.
- [Portfolio Visualizer alternative: when to switch and when not to](https://www.ensemble.markets/compare/portfolio-visualizer-alternative): An honest comparison of Portfolio Visualizer and ENSEMBLE for portfolio backtesting: what Portfolio Visualizer does well, its pricing and limits as of September 2026, what it cannot express, and the cases where each tool is the better choice.
- [testfol.io alternative: an honest comparison with ENSEMBLE](https://www.ensemble.markets/compare/testfolio-alternative): testfol.io is the community favorite for fast, free portfolio backtests on daily data. This comparison covers what it does well, its pricing as of September 2026, where its form-based approach stops, and when ENSEMBLE's plain-language strategies, composition, and agent access are the better fit.

## Model portfolios for advisors

- [For advisors](https://www.ensemble.markets/advisors): Build, backtest, document, and keep model portfolios current from plain language, with white-label fact sheets and allocation files for Altruist and Interactive Brokers.

## Optional

- [Privacy](https://www.ensemble.markets/privacy)
- [Terms](https://www.ensemble.markets/terms)
- Support: support@ensemble.markets