Ask what you’re trying to understand. Research the market, read current conditions, evaluate a contract, run an optimization and interrogate the model evidence, all inside one continuing session.
Start with the question you actually have. DeriveAI reads what you are asking for and brings the right intelligence into the session, rather than making you choose between a screener, a chain and a model output before you have asked anything.
One session instead of ten tabs
Research usually means moving between a screener, a chain, a chart and three dashboards that do not know about each other. A session keeps the market read, the candidates, the contract evaluation and your own history in one place, in the order you asked for them.
It can see the models
You can ask about model output as easily as about a price: what the odds were, which model version produced them, and how that model has been holding up. Every answer carries what produced it.
What a session looks like
Four questions, four different kinds of intelligence, one thread. Nothing here asks you to know which part of DeriveAI to open, and nothing is presented without saying where it came from.
How does the market look today?
Routed to
Market Read
Selective · Bullish
What looks interesting?
Routed to
Discovery
Candidates worth a closer look
Review this NVDA contract.
Routed to
FTO-1
A probability, and the calibration behind it
Optimize NVDA under $500.
Routed to
Optimizer
Eligible contracts, ranked by their odds
An example session. Reads and rankings are whatever the market and the models return on the day.
Where it sits
The workspace is where the decision happens. Market Read, Discovery, Optimizer and FTO are capabilities it calls when your question needs them, not places you go first.