Skip to content
Technology How it works Breeder — Hyperion Burner — Aegis Burner — MetroVolt AI-Native Architecture Magnets Fuel cycle Safety Roadmap
Solutions AI & Data Centers Defense & Government Grid & Baseload Neutron Detection Quantum
Learn Technical Library
Proof Publications Whitepapers Technical Library Open Science & Reproducibility The Honest Gates
Company About / Mission Leadership Environment Health & Safety Investors Careers Press Contact
3D Model
AI Architecture › L4 · Orchestration
L4 · Orchestration

Deterministic Replay

Any shot or campaign can be re-executed against the recorded event stream to reproduce exactly what orchestration saw and decided.

THE STACK · click to jumpL7Ecosystem & StrategyL6Experience & VisualizationL5Applications & CopilotsL4OrchestrationL3Twin Modeling & AIL2Data FabricL1Control PlaneL0Foundation▲tlmctl▼L4 · ORCHESTRATIONEvents, workflows, rules, and human routing.1Event Streamingthe backbone2Workflow Enginecampaign procedures3Rules & Safety Boundshard limits4Human-in-the-Loopapproval routing5Schedulerexperiment campaigns6Audit Busfull decision lineageMACHINE TIECoordinates L3 outputs with L5 copilots and human operators.KRONOS FUSION ENERGYAI-NATIVE S.M.A.R.T. GENERATORORCHESTRATIONSHEET 06REV. 2026-08L4 · AI-NATIVE STACK
L4 · Orchestration — its place in the stack (left, click any layer) and its internal components (right). Telemetry rises; control descends.

Replay as ground truth

Because the event log is the single source of truth and procedure logic is deterministic, orchestration can re-run any past shot bit-for-bit. Feed the recorded events back into the same procedure version and it reaches the same states and would emit the same commands. This is the core tool for post-shot analysis on the breeder and for validating orchestration changes before they touch either machine.

Requirements for determinism

python
def replay(shot_id, procedure_version):
    events = log.read(correlation_id=shot_id)   # exact recorded stream
    engine = Engine(procedure_version, mode=REPLAY)  # commands are captured, not sent
    for e in events:
        engine.apply(e)
    return engine.emitted_commands, engine.final_state

# assert replayed commands == originally logged commands  (regression guard)

What-if replay

A variant, covered in replay and what-if simulation, swaps in a new procedure or twin version and replays historical inputs to ask: would the change have behaved better on last week's breeder disruption? Emitted commands are captured, never sent, so the machine is untouched.

Auditable causality

Replay is bound to the decision-lineage bus: causation and correlation IDs let an investigator reconstruct the full chain from a raw diagnostic sample through a twin prediction and a copilot proposal to the gated command that moved a burner plug magnet. Determinism turns 'why did it do that' from a guess into a reproducible experiment.

On both machines this discipline is validated today against synthetic and simulated campaigns, long before FOAK first tritium around 2030.

Content reviewed August 2026 · design-and-simulation stage