Quant Development (C++ and Python)

Locked Python environments, a disciplined C++ toolchain, numerical reference tests, stored benchmarks and profiling habits, plus the review rota and release routine that keep a research team's quant library fast and correct.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Then open the Marketplace in the app, search for Quant Development (C++ and Python), and click Install. It goes into your own files, and works offline from then on.

Before you start: this plan is general information to help you organise decisions about money. It is not financial, investment, tax or legal advice, and it does not recommend any product, investment or deal. Rules differ between countries, so check decisions that matter with a qualified, licensed adviser in yours. Investing and trading can lose money, and raising or managing other people's money is regulated in most countries.

What it installs.

1

Pillar

1

Area

50

Projects

200

Starter tasks

🧮 Quantitative Finance & Trading

Quant Development (C++ and Python)

Writing production-quality quant code: Python research stacks, C++ for performance, testing numerical code, profiling and code review for research teams.

  1. Reproducible Python research environment with locked dependencies
  2. C++ toolchain, compiler flags and CMake baseline
  3. Inventory of quant libraries, owners and callers
  4. Continuous integration building Python and C++ together
  5. Numerical reference test suite for core analytics
  6. Coding standard for numerical C++ and Python
  7. Benchmark harness with stored baselines
  8. Precision, dtype and tolerance policy
  9. Developer workstation and remote build setup
  10. Weekly performance regression check
  11. Monthly dependency and compiler upgrade window
  12. Review rota for research pull requests
  13. Scheduled sanitizer and static analysis runs
  14. Flaky and slow test triage
  15. Release tagging and changelog for the quant library
  16. Quarterly technical debt review with researchers
  17. Production numerics incident log
  18. Developer support hours for researchers
  19. Monthly build time and CI cost review
  20. Vectorised NumPy and pandas idioms for research code
  21. Polars and Arrow for larger research data sets
  22. Numba and Cython for hot loops
  23. Modern C++ for numerical work
  24. Python bindings with pybind11 or nanobind
  25. Floating point error, cancellation and summation
  26. Profiling Python and C++ with sampling profilers
  27. Memory layout, cache behaviour and SIMD basics
  28. Processes, threads and the GIL in research code
  29. Porting a slow Python analytic into native code
  30. Deciding which code moves from Python into C++ and why
  31. Choosing linear algebra, RNG and numerical libraries
  32. Reproducible random number scheme across the library
  33. Property-based tests for pricing invariants
  34. Packaging the quant library as binary wheels
  35. Consolidating duplicated date and calendar utilities
  36. Retiring a dead or duplicated numerical routine
  37. Python or compiler major version migration
  38. First stable API release of the quant library
  39. Performance sprint on the overnight batch
  40. Production readiness review of research model code
  41. Post-incident review after a numerical bug reached production
  42. First 90 days as a quant developer on a research team
  43. Moving from software engineering into quant development
  44. Sole developer supporting a small fund's researchers
  45. Leading a small quant development team
  46. Handover pack before changing desks or employers
  47. Personal open-source quant code portfolio
  48. Automatic differentiation for Greeks and calibration
  49. GPU acceleration for Monte Carlo pricing
  50. Bit-reproducible results across machines and builds

Where it starts.

Each project arrives with its purpose, milestones and first tasks written out. Here is how the first three read.

Reproducible Python research environment with locked dependencies

Researchers who rerun last quarter's notebook and get a different Sharpe ratio have usually been bitten by an unpinned NumPy, pandas or SciPy upgrade, not by the market. Locking every dependency, including the BLAS build underneath NumPy, means the same commit gives the same numbers on any machine and a new joiner is productive on day one.

First tasks

  • Export the packages and versions currently installed on your research machine
  • Choose one environment manager and write the reason in the repository readme
  • Generate and commit a lockfile that pins the BLAS backend as well
  • Rebuild on a clean container and compare a reference notebook's output byte for byte

C++ toolchain, compiler flags and CMake baseline

Quant C++ that builds with one developer's flags and crashes with another's is a common source of results nobody can reproduce, especially when -ffast-math or a different -march slips into a release build. A single CMake baseline with agreed compilers, language standard, warning levels and optimisation flags makes debug, release and sanitizer builds behave predictably everywhere.

First tasks

  • List the compilers, versions and flags each developer currently builds with
  • Write a CMake presets file with debug, release and sanitizer configurations
  • Turn warnings into errors in CI and fix or document the existing ones
  • Record the team decision on fast-math and target architecture flags

Inventory of quant libraries, owners and callers

Most research teams run on a few shared libraries nobody fully owns: a curve builder, a calendar module, a pricing core and a dozen utility files copied between repositories. Listing each one with its language, owner, callers and test coverage shows where a bug would do the most damage and where duplicated code is quietly giving two different answers.

First tasks

  • Search the repositories for modules imported by more than one project
  • Fill in language, owner and main callers for each library found
  • Mark any calculation implemented twice in different places
  • Refresh the library inventory and its owners @recurring(yearly)

Make it yours.

Install Quant Development (C++ and Python), then tell Aurum the date you're aiming for. It checks whether the plan fits the weeks you actually have.

Free · Mac (Apple silicon) and Windows · v1.0.5 · Other platforms

Blueprints are general information to help you organise your own work. They are not medical, financial, legal or other professional advice. See the Terms.