What Is Quant Mode?

Quant Mode is PyP's Python-based strategy workflow for quantitative traders, prop researchers, and systematic signal builders.

It exists for the cases where a .ppc strategy language is not the right abstraction and you want to work directly with Python code, model artifacts, and deployment-ready signal contracts.

Quant Mode is not a notebook host and it is not a generic backtesting sandbox. It is a structured pipeline:

1. Write a strategy in strategy.py
2. Define runtime and training configuration in quant.config.json
3. Launch a quant job against PyP market data
4. Produce a versioned artifact
5. Validate the release with simulation and PPE / PPE-PP
6. Deploy live signals to channels and terminals
7. Publish verified releases to the marketplace

When to use Quant Mode

Use Quant Mode when:

• you want to write plain Python instead of PyPScript
• your strategy is model-based rather than rule-language-based
• you need train(data, config) and predict(model, market_data, config) directly
• you care about artifacts, releases, and live runtime behavior as first-class platform concepts

Use PyPScript when:

• you want a readable rule language for pattern-based strategies
• you are expressing deterministic market logic rather than training a model artifact
• the strategy should stay inside PyP's native strategy-authoring abstraction

The Quant project shell

Quant projects live under a dedicated shell:

• /projects/quant/new
• /<user>/<project>/quant/research
• /<user>/<project>/quant/datasets
• /<user>/<project>/quant/jobs
• /<user>/<project>/quant/artifacts
• /<user>/<project>/quant/releases
• /<user>/<project>/quant/simulations
• /<user>/<project>/quant/deploy

That separation is deliberate. Quant should feel like its own technical workflow, not like a few extra controls bolted onto the PyPScript editor.

What makes Quant Mode different

Quant Mode is built around:

• Python strategy code
• explicit model artifacts
• release state
• simulation-backed publish gating
• runtime-aware deployment

The deployable contract is not "some code ran once." The deployable contract is a release backed by an artifact and validated through the same downstream lifecycle PyP uses for real signal products.

Recommended reading order

1. Research Workspace
2. Strategy Contract
3. quant.config.json
4. Training Jobs
5. Artifacts & Model Formats
6. Releases & Simulations
7. Live Deployment