Define the research case
Record the universe, hypothesis, period, benchmark, data source, costs and execution assumptions.
From a backtest run to an investment decision.
The upcoming QuantJourney Research Workspace will connect strategy setup, execution evidence, validation and review in one auditable workflow. The open engine is available today; hosted team access is not public yet.
Available now: Python engine and reports Coming soon: hosted research workspace
One research record
Built for the work after the chart looks good.
The workspace is designed to preserve the decisions behind a result: the exact strategy source, data assumptions, execution path, costs, validation evidence, logs and review notes.
Run with live evidence
Follow deterministic run phases, logs, warnings and engine state without losing the active run.
Review the full result
Inspect performance, risk, trades, orders, fills, metadata and source captured for that run.
Compare and decide
Compare like-for-like runs and assemble a research packet for PM, risk or investment review.
Governance by design
Institutional research controls, planned from day one.
Most backtesting stacks stop at the equity curve. The workspace is being designed around what an investment process needs afterwards: provenance, separation of duties, statistical evidence and an audit trail that survives personnel and time. Planned capabilities — subject to change before general availability.
Immutable run records
Every run preserves its strategy source, configuration, seeds, data snapshot reference and outputs as one referenceable record — the result and the assumptions behind it stay attached.
Data provenance and lineage
Runs reference versioned data snapshots with provider lineage, so any result can be traced back to the exact inputs that produced it — months later, by someone who was not in the room.
Statistical validation gates
Permutation tests, Deflated Sharpe, explicitly labeled rank stability and walk-forward evidence attach to the run record itself, not to a slide deck — a strategy advances only with its evidence.
Roles and access control
Research, review and read-only roles keep authorship and approval separate. The person who built a strategy is not the one who signs it off.
Review workflows
Comments, checklists and sign-offs happen on the run record, producing a review trail that investment committees and internal audit can rely on.
Research packets
Exportable evidence packets — metrics, plots, assumptions, validation results and logs — in a reviewable format ready for PM, risk or client due-diligence conversations.
Who it serves
One record, three audiences.
The same run record answers different questions for different desks — without anyone re-running notebooks or reconstructing assumptions from memory.
Portfolio managers
Compare candidate strategies like-for-like and see the assumptions and validation behind every equity curve before allocating.
Quant research teams
One explicit standard for how strategies are configured, run, validated and reviewed — consistent across people, desks and time.
Risk and compliance
Evidence retention, reproducibility and separation of duties designed to fit internal research-governance and model-review requirements.
Not publicly open yet.
We are finishing access controls, data provenance, run history and operational capacity before opening the hosted workspace more broadly. The open-source engine is available today and the workspace will build on the same run artifacts.