CONTINIUMMTSM · Simulation Engine
Thousands of futures. Fast enough to iterate.
MTSM is the modeling core: it runs high-volume simulations so you can explore decisions at the speed of thought — then refine assumptions until the result holds up.
DEFAULT MODE
Clarity-first
Explanations over vibes.
OUTPUT
Future Snapshot
Narrative + deltas.
QUICK LOOK
Forks, not a single path
compare outcomes
Assumptions visible
edit + rerun
Explainable deltas
why it changed
RUN A BATCH
POST /api/simulations/generate
Content-Type: application/json
{
"prompt": "If I change roles, how does my next year change?",
"horizonMonths": 12,
"variants": 64,
"assumptions": {
"salaryGrowth": 0.06,
"burnRateUSD": 2800
}
}Think in batches: sweep assumptions and compare distributions, not single-point outcomes.
OVERVIEW
Designed for decisions you can defend
The core idea is simple: forks stay comparable, assumptions stay visible, and outputs stay legible.
01
High-throughput runs
Explore wide scenario spaces quickly so you can ask better questions earlier.
02
Model transparency
Clear inputs and assumptions, with outputs that map back to the decision that created them.
03
Stable comparisons
Designed for side-by-side deltas: what changed between A and B, and what stayed constant.
CAPABILITIES
A small surface area. Deep leverage.
Capabilities are designed to compose: orchestrate → simulate → compare → narrate.
Parameter sweeps
parameterSweeps()
Vary salary growth, burn, health routines, risk, and more to see sensitivity and robustness.
Scenario sampling
scenarioSampling()
Generates diverse candidate futures while keeping the decision framing consistent.
Outcome aggregation
outcomeAggregation()
Summarizes distributions and tails — not just averages — so you can see risk.
Composable inputs
composableInputs()
Works with TIE, TCE, and the Dynamic Adapter to keep a single source of truth.
OUTPUTS
Artifacts you can inspect
Outputs are structured so you can compare forks cleanly, trace drivers, and keep context over time.
ARTIFACTS
• Outcome distributions (not just single-point predictions)
• Scenario clusters and representative futures
• Sensitivity hints: which assumptions matter most
• Structured outputs ready for narrative hydrationThink of these as the stable, comparable units the UI (and you) can reason about.
NEXT STEP
Simulate first. Optimize second.
MTSM gives you a realistic, testable foundation for decision-making — then the rest of the stack turns it into clarity and narrative.
No commitment. Just insight.