Skip to content
PERYX.AI

Products

One architecture. One industry at a time.

Peryx goes deep on a single industry's physics, operations, and failure modes before calling a platform done. Everything below carries its real status.

Available today

Available today

Solar Intelligence Platform

Physics-calibrated performance intelligence for solar fleets — every megawatt-hour of loss attributed to a cause.

Explore the platform
  • Every five-minute interval compared against a model of your own plant, built from its PAN/OND configuration
  • Soiling, string mismatch, inverter drift, and equipment failure separated and quantified independently
  • Heterogeneous inverter brands, module vintages, and sites normalised into one fleet picture

Roadmap

Stated plainly, because an unlabelled roadmap becomes a promise.

  • In development

    Digital Twin Engine

    The modeling core underneath every Peryx platform, being generalised beyond solar assets.

    • — Asset models calibrated from manufacturer specifications and observed behaviour
    • — Continuous reconciliation between modeled and measured state
    • — Deviation attribution rather than threshold alarms
  • In development

    Enterprise AI APIs

    Programmatic access to telemetry, model outputs, and derived intelligence for your own systems.

    • — Query normalised telemetry across sites and vendors
    • — Pull model outputs, attributed losses, and classified events
    • — Push intelligence into the tools your operations team already uses
  • Planned

    Utility Intelligence

    Grid and distribution asset intelligence for utilities and network operators.

    • — Network-level asset health from existing SCADA and metering data
    • — Load and constraint behaviour modeled per asset class
    • — Intervention prioritised by consequence, not by alarm count
  • Planned

    Manufacturing Intelligence

    Process and equipment intelligence for asset-heavy production lines.

    • — Process lines and rotating equipment modeled against design intent
    • — Drift detected before it becomes scrap or downtime
    • — Maintenance sequenced by predicted consequence
Available today
In production use. You can deploy it now.
In development
Being built now, with a working internal implementation.
Planned
Architecture designed, scheduled for build.

The shared substrate

Every product sits on the same seven stages.

This is why a second industry is a modeling problem rather than a rebuild — the ingestion, twin, and attribution layers do not change.

  1. Sensors

    What the asset already reports

  2. Telemetry

    Ingested at native resolution

  3. Physics model

    What this asset should be doing

  4. Digital twin

    Continuously reconciled

  5. Insights

    Deviation, attributed

  6. Recommendations

    Ranked by consequence

  7. Decisions

    Acted on in your systems

How the architecture works

See what your operation is actually losing.

A walkthrough on your own asset class, with a scoped pilot on one to three sites if it's a fit.