I am building a modular research and back-testing environment that lets a non-programmer trader concentrate solely on entry and exit rules while every other task—data handling, trade simulation, capital allocation, optimisation, reporting and portfolio analytics—runs automatically in the background. Scope of the platform • Asset coverage must include equities, futures, options, commodities and crypto from day one. • Users will enjoy advanced, unrestricted control over the trading logic itself. All surrounding infrastructure must stay “black-box” to them so they can’t accidentally break core processes. • During Phase 1 the engine should pull, clean and align historical market data. Phase 2 will extend the same framework to live feeds without rewriting user strategies. • Trading logic will be entered through straightforward Python snippets: if a user can write a few lines of condition/-action code, the system should compile, back-test and report results automatically. No other programming knowledge should be required. Key modules to deliver 1. Data manager: ingestion, storage, resampling, corporate-action adjustment. 2. Execution simulator: realistic fills, slippage, margin, commissions. 3. Portfolio engine: capital allocation, risk limits, multi-asset netting. 4. Optimiser: walk-forward, grid and heuristic search with out-of-sample validation. 5. Report suite: equity curve, tearsheets, factor exposure, trade logs and downloadable CSV/JSON. 6. Plug-in layer: allows me to drop in new asset classes, data vendors or brokerage APIs later with minimal refactor. Acceptance criteria • A sample momentum and a delta-neutral options strategy run end-to-end with one command. • Results between repeated runs on identical data sets are deterministic. • Clear developer and end-user documentation, plus screencast or notebook showing strategy creation, test run and report interpretation. Tech stack is flexible, but Python (Pandas, NumPy, vectorised back-test tools, Plotly/Matplotlib) feels natural; feel free to suggest faster cores in Cython/Rust if they stay transparent to the end user. If this sounds like your domain, outline the architecture you’d follow, any comparable frameworks you have built, and an estimated timeline for Phase 1 deliverables.