About NOVAQ
An independent project that set out to build a complete research and trading system around one question: what is actually true in trading, and how would we know?
Origin
NOVAQ started on 14 May 2026 as an attempt to make years of trading education usable. The starting point was practical: a large body of trading knowledge that could not be searched, compared or tested. Solving that called for a system to store knowledge exactly, a laboratory to test it, a platform to act on it and audit tools to check it all. The project grew in that order.
What was built
In under six months, one operator working with AI agents built about 145,500 atomic knowledge notes, 16 years of annotated NQ and ES session history, about 700 catalogued studies and tests, 161 trading legs, a platform of nine supervised service groups with about 4,900 automated test functions, and an eight-role AI audit system. Each part works on its own, and the value of the project lies in their connection: a claim from a book can become a test, a test can become a strategy, and a strategy's live behaviour can be compared with the research behind it.
What drives the project
Trading knowledge is abundant, and tested knowledge is scarce. NOVAQ exists to close that gap with engineering: structured knowledge, clean data, strict testing and a system that carries results through to execution.
Four principles run through the work.
- Attributed. Each claim traces to an author and a source.
- Point in time. A decision may use only what was known at that moment.
- Reproducible. A result can be rebuilt from its recorded inputs.
- Faithful. Ports and translations reproduce earlier behaviour exactly, and improvements come afterwards.
The role of AI
NOVAQ is run by one operator with AI agents. AI multiplies output, so the project invests equally in the tools that verify it.
Direction
- Continue neural research, evaluated strictly on point-in-time inputs.
- Extend position sizing to account limits.
- Carry research candidates through forward testing.
- Connect the growing knowledge base more closely to research.
These are intentions, not commitments.
Collaboration
Part of NOVAQ's research is carried out jointly with Vašek (BTE). Vašek brings studies on 16 years of NQ and ES data with pre-registered tests, an out-of-sample period and correction for multiple testing. NOVAQ reproduces them independently on its own data and checks whether a result can become a tradable strategy after costs. The findings then feed the knowledge base.
Joint work has produced, for example, studies of auction statistics, options positioning and open types. They are published in a generalised form, without the definitions and parameters needed to repeat them.
NOVAQ does not offer signals, managed accounts or investment products. It has no other team members or commercial partners to announce at this time.
Contact
Author of the NOVAQ system: Tibor
Email: novaq@novaq.sk
Questions about the project and its research are welcome.