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

The Engineering Copilot: Scope and Reasoning Loop

An advisory agent for subsystem analysis, anomaly triage, and maintenance reasoning across magnets, vessel, fueling, and the DEC train.

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.

What the Engineering Copilot does

The Engineering Copilot is the hardware-facing agent of L5. Where the Plasma Copilot reasons about the plasma, this one reasons about the machine: superconducting magnets, the vacuum vessel and plasma-facing components, cryogenics, power supplies, the fuel cycle, and — on the burner — the direct-energy-conversion train. It performs subsystem analysis, triages anomalies raised by L3, and plans maintenance.

Three capabilities

Like all L5 agents it grounds every claim with retrieval over the data fabric — design documents, prior fault records, maintenance logs, and vendor specifications — and it proposes rather than acts. A maintenance action or a subsystem setpoint change is a proposal to L4, not a command.

Both machines' subsystems

Engineering Copilot subsystem focus
MagnetsREBCO strain, quench precursors, delta-T26.49 T plug / 17 T throat coilsVessel / PFCneutron fluence, thermal load from 85.0 MWexpander and plug thermal loadFuel cycletritium (~4 kg/yr) + He-3 (~1.97 kg/yr) balanceD-3He supply, He-3 handlingConversionthermal balance-of-plantDEC collector-grid health

The reasoning loop mirrors the Plasma Copilot's: assemble context from live twin state and retrieval, plan bounded read/simulate tool calls, iterate, and compose an answer with citations and uncertainty. The subsequent pages detail subsystem analysis, anomaly triage, and maintenance reasoning. Because both machines are design-and-simulation studies, the copilot is exercised today against the twin and the commissioning plan; its maintenance reasoning targets FOAK readiness rather than an operating fleet.

Content reviewed August 2026 · design-and-simulation stage