TRADING INTELLIGENCE PLATFORM
Discover what the market can reveal.
Kyro brings market data, research, strategy discovery, testing and evidence into one platform — so a trading idea has to earn its place before it ever reaches a decision.
Kyro states probabilities, never promises. It holds no broker credentials and cannot place a trade.
Evidence before conviction
HOW AN IDEA EARNS ITS PLACE
- E0Idea
- E1Published
- E2Replicated
- E3Robust
- E4Shadow
- E5Live
- E6In use
WHAT THE PLATFORM IS MADE OF
- Data
- Every source validated and timestamped
- Research
- Questions asked against the whole record
- Strategy discovery
- Tested, then promoted — or rejected
- Market intelligence
- Structure, liquidity, sessions, crowd
- Decision
- A probability with its reasoning attached
- Learning
- Did the stated confidence hold up?
THE PLATFORM
Five things Kyro does, in the order they matter
Each one is a real part of the architecture. Where a capability is still being built, Kyro says so rather than implying otherwise.
Discover
Scan markets and instruments for what is worth a closer look, rather than watching one chart and hoping.
In buildResearch
Investigate how markets actually behave, against sources that can be cited back to the page or the minute.
WorkingTest
Evaluate an idea against history with costs and assumptions stated, and the number of variants tried declared.
WorkingCompare
Put strategies side by side on the same evaluation. When they disagree, see the disagreement.
In buildLearn
Turn finished research into knowledge the platform keeps — including the ideas that stopped working.
Working
WHY KYRO EXISTS
Don't trade on assumptions. Test them.
Most trading ideas are never measured. They are remembered when they work and forgotten when they don't. Kyro is built the other way round — every claim carries the record that produced it, and an idea that cannot survive being tested does not get to influence a decision.
- Every claim carries its record
- Source, version and timestamp — openable, not implied.
- The honest score is the headline
- Corrected for how many variants were tried, with the count shown.
- Disagreement is information
- Two strategies pointing opposite ways is shown, not averaged away.
- A dead idea stays visible
- Retired rather than deleted, so it cannot return looking new.
- Confidence is measured, not asserted
- If Kyro says 84%, that number is checked against what happened.
THE DIRECTION
One trader, one strategy, one instrument is not the ceiling
Kyro is being built to investigate a broad strategy universe and work out what is genuinely worth pursuing. This is the direction of the platform, not a description of a finished library.
Trend
E1Momentum and breakout behaviour, studied across a century of data
Carry
E1A long-documented effect in currency markets
Microstructure
E2Where resting orders cluster, and what happens when they are reached
Sessions
E2Among the most repeatedly confirmed patterns in intraday markets
Mean reversion
E1Short-horizon reversal, well covered in the literature
Positioning
E1What the crowd is doing, as a signal in its own right
Volatility
E1Useful for sizing and stops rather than direction
Ensembles
E1Combining many weak predictors instead of trusting one
Grades describe how well documented a family of ideas is, not how profitable it would be. A well-documented effect still has to survive Kyro's own testing before it is used.
METHODOLOGY
Research. Test. Measure. Repeat.
Kyro evaluates ideas against real market data with explicit assumptions and measurable outcomes. The uncomfortable parts are shown rather than smoothed over.
01
Research
Pose a question the record can actually answer.
02
Test
Run it against periods it has never seen, costs included.
03
Measure
Compare what was expected against what happened.
04
Repeat
Keep what survives. Retire what does not, in public.
Start with the evidence.
Kyro is in active development. What exists is real, what doesn't is named — on the platform itself.