# ENSEMBLE

> Build, test, and maintain investment models with your AI agent. ENSEMBLE provides the data, research tools, and infrastructure to keep them running. Inspect asset coverage, model code, inputs, builds, runs and current allocations. Backtests are hypothetical research.

## For AI agents

- [MCP server](https://api.ensemble.markets/mcp): remote Streamable HTTP, OAuth or bearer API key. Discover tools for assets, models, runs, builds, backfills and optimizers from the connected server; do not rely on a fixed tool count.
- [ENSEMBLE in the Claude directory](https://claude.ai/directory/ensemble): choose Connect to Claude, sign in to ENSEMBLE, and confirm the connected account before making changes. Plan and workspace access requirements apply.
- [REST API](https://api.ensemble.markets/v1): canonical product surface. [OpenAPI](https://api.ensemble.markets/v1/openapi.json), [reference](https://api.ensemble.markets/docs). Lists use data/total; read variants use include; errors use a shared envelope.
- [CLI](https://www.ensemble.markets/cli): npx --yes --package=https://www.ensemble.markets/downloads/ensemble.markets-3.3.0.tgz ensemble <noun> <verb>; use --help and ensemble schema for discovery.
- [Pricing as markdown](https://www.ensemble.markets/pricing.md): current plans, credit terms, capabilities and metered rates.

Models can declare an ordered universe input containing ticker symbols and owned model IDs. Change it per run through the published inputs_schema; a successful promotion updates live defaults and dependencies. Legacy fixed-universe scripts remain supported.

## Product

- [Asset screener](https://www.ensemble.markets/assets): exposure, structure, tax classification, leverage and available price history. Unknown data stays explicit.
- [Backfills](https://www.ensemble.markets/assets#backfills): extend histories with visible assumptions; synthetic history is distinct from observed history.
- [Models](https://www.ensemble.markets/models): natural-language or supplied-code construction, custom inputs, composition, stored builds and runs.
- [Optimizers](https://www.ensemble.markets/models#optimizers): store and reuse custom allocation logic.
- [Web app](https://www.ensemble.markets/gui): the available browser workspace, with an interactive local sample before signup.
- [Pricing](https://www.ensemble.markets/pricing): subscription tiers plus metered execution, with account spending caps.
- [MCP setup](https://www.ensemble.markets/mcp): connection and authentication instructions.
- [Methodology](https://www.ensemble.markets/methodology): data limits, runtime assumptions, research records and execution responsibilities.

Scheduled model refreshes depend on readiness, data availability and account limits. Inspect the latest successful run and data date. Broker-format files require Business; they do not place trades or establish broker tradability. Stored records do not imply immutable snapshots of every data dependency.

## 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): Define relative and absolute momentum separately, inspect a public model, and compare a parameter change. The example explains the research workflow without promising a historical or future result.
- [Risk parity: how it works, how to size it, and a backtest](https://www.ensemble.markets/strategies/risk-parity): Understand risk contributions and the simpler inverse-volatility weighting rule. Inspect a public model, test one assumption, and distinguish a model specification from a performance claim.
- [The 60/40 portfolio: rules, variants, and a backtest](https://www.ensemble.markets/strategies/60-40-portfolio): Define a stock-and-bond allocation, inspect a public 60/40 model, and test rebalancing or asset changes over matched dates. Learn what the model shows and what it leaves out.
- [The 200-day moving average strategy: rules, evidence, and a backtest](https://www.ensemble.markets/strategies/200-day-moving-average): Define a moving-average allocation rule, specify signal and execution timing, and inspect its tradeoffs in a model run. An illustrative prompt helps you build and compare your own version.

## Portfolio backtesting software for your own investment models

- [Backtesting](https://www.ensemble.markets/backtesting): Build and backtest your portfolio in ENSEMBLE. Inspect the code, historical results, and allocations. Start with $100 in metered credit, valid for 7 days; no card required.
- [How to backtest a portfolio without writing code](https://www.ensemble.markets/backtesting/how-to-backtest-a-portfolio): Research two assets, build an allocation model, inspect its saved results, and compare one change. A practical first workflow for DIY investors using ENSEMBLE.
- [Backtest metrics explained: CAGR, volatility, Sharpe ratio, and max drawdown](https://www.ensemble.markets/backtesting/metrics-explained): Read ENSEMBLE backtest metrics with their actual conventions: CAGR, annualized volatility, a zero-risk-free Sharpe calculation, maximum drawdown and trading activity. Compare matched windows and inspect assumptions.
- [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): Research assets and build, test, and maintain allocation models with a compatible AI client. Inspect saved code, inputs, runs, and current allocations.
- [Build and maintain investment models with an AI agent](https://www.ensemble.markets/agents/build-and-maintain-investment-models): Use current ENSEMBLE tools to research assets, create a saved model, inspect a run, compare inputs and maintain reviewed allocations through explicit refreshes.
- [A finance MCP server for allocation models and research](https://www.ensemble.markets/agents/finance-mcp-server): Use ENSEMBLE's hosted MCP server to research available assets, build allocation models, inspect saved runs, and retrieve allocations. Understand connection requirements, data coverage, and metered execution before starting.
- [How to research and backtest a model with Claude](https://www.ensemble.markets/agents/backtest-with-claude): Connect a compatible Claude client to ENSEMBLE, verify asset coverage, build an allocation model, and inspect a parameter test. An illustrative workflow with current tool names and clear execution limits.

## 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.
- [A Portfolio Visualizer alternative for your own investment models](https://www.ensemble.markets/compare/portfolio-visualizer-alternative): Build custom allocation models with ENSEMBLE: inspectable code, saved backtests, and AI agent access. Try $100 in metered credit for 7 days; no card required.
- [testfol.io and ENSEMBLE: choosing a research workflow](https://www.ensemble.markets/compare/testfolio-alternative): Compare testfol.io's portfolio research tools with ENSEMBLE's stored allocation models, custom Python, model composition, and connected workflows. Check current features and costs before choosing.

## Model portfolios for advisors

- [For advisors](https://www.ensemble.markets/advisors): Build, backtest, document, and keep model portfolios current from plain language, with Business features for white-label fact sheets and supported broker-format allocation files.
- [Export reviewed model allocations for your brokerage workflow](https://www.ensemble.markets/advisors/export-model-allocations): Inspect a model's run and target weights, select a supported Business export, and review the file before using it with Altruist or Interactive Brokers.

## Optional

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