Skip to content
All library documents

AI-Assisted Refactoring of a Multi-Account VeighNa Trading Platform

Article vn.py community
Read the original

Summary

The author describes using an AI assistant to rebuild a customized trading application from VeighNa 3.9 to 4.3. The work covered a changed modular architecture, a new database backend, a redesigned strategy data model, and a rewritten desktop interface. The reported process relied on the human to define requirements, review code, and make decisions while the AI produced the implementation over roughly a week; this is a project account, not a controlled evaluation of AI coding quality.

The resulting application is described as supporting automatic and assisted trading modes, multiple live and simulated accounts, contract roll handling with position protection, strategy parameter and state displays, performance views, and batch market-data import. An experimental AI feature is intended to help interpret strategy signals. The article does not describe specific trading strategies, report trading results, or provide benchmarks for reliability, execution, or the AI feature. Its lessons concern software architecture and workflow for a trading system rather than evidence that the system or strategies improve returns.

Key ideas

  • The author used AI to help rebuild a customized trading application across major VeighNa versions.
  • The described workflow assigns requirements and review to the human while AI generates implementation code.
  • The redesigned platform supports multiple accounts and automatic or assisted trading modes.
  • The strategy interface separates parameters, state, and internal variables.
  • The article reports project features but gives no trading performance or reliability benchmarks.

Tags

ExecutionFuturesBacktestingMachine learning

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.