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 › L5 · Applications & Copilots
L5 · Applications & Copilots

Operations Copilot: Commissioning and First-Campaign Support

Generating and sequencing the commissioning procedures that take the breeder from construction start (Q2 2027) toward FOAK first tritium (~2030).

THE STACK · click to jumpL7Ecosystem & StrategyL6Experience & VisualizationL5Applications & CopilotsL4OrchestrationL3Twin Modeling & AIL2Data FabricL1Control PlaneL0Foundation▲tlmctl▼L5 · APPLICATIONS & COPILOTSAgentic copilots that reason over the machine.1Plasma Copilotscenario design2Engineering Copilotsubsystem analysis3Operations Copilotrunbooks & procedures4Agentic Toolsbounded action-taking5Knowledge BaseRAG over the fabric6Guardrailssafety-boundedMACHINE TIEReads the twin and fabric; proposes actions that route through L4.KRONOS FUSION ENERGYAI-NATIVE S.M.A.R.T. GENERATORAPPLICATIONS & COPILOTSSHEET 07REV. 2026-08L5 · AI-NATIVE STACK
L5 · Applications & Copilots — its place in the stack (left, click any layer) and its internal components (right). Telemetry rises; control descends.

Commissioning is the near-term product

Both Kronos machines are design-and-simulation studies; the breeder (Hyperion) begins construction in Q2 2027 with first-of-a-kind first tritium targeted around 2030. The nearest-term work the Operations Copilot supports is therefore commissioning: the ordered, gated sequence of tests that brings each subsystem online and validates it against the twin before the next is enabled.

Gated commissioning sequence

Commissioning proceeds through hardware-in-the-loop and staged energization gates. The copilot generates the procedure for each gate, defines the pass criteria against twin predictions, and holds progression until the prior gate is signed off by a human. It cannot advance a gate on its own.

text
commission(gate):
  proc = generate_procedure(gate.kind, 'breeder')
  pass_criteria = twin.predict(gate) +/- tolerance
  run under hardware-in-the-loop
  if measured within pass_criteria AND human_signoff:
     enable(next_gate)
  else: hold, raise to Engineering Copilot for analysis

Twin as the reference

Every commissioning measurement is compared to the twin's prediction; a mismatch is a finding routed to the Engineering Copilot, not a number to be explained away. This is how the twin is validated against the real machine as it comes online, and how the copilots earn calibrated confidence before they advise on real shots. No performance or net-gain claim is made during commissioning — the frozen design point (Q_sci 3.076, 85.0 MW) is a target the built machine must demonstrate.

For the burner (Aegis / MetroVolt) the same commissioning machinery applies to plug energization (26.49 T), ambipolar-potential establishment, and DEC-train bring-up on its own later schedule. See hardware-in-the-loop commissioning and honest gates.

Content reviewed August 2026 · design-and-simulation stage