Skip to content

Taranis

Anticipates blackouts and grid instability before they happen.

Vertical: Energy Solid base

What it is for

Prevent blackouts by anticipating when and where the power grid will become unstable, before it does.

What it does

Continuously observes grid state, simulates its behaviour in a digital twin, detects patterns that preceded real blackouts, analyses what would happen if one or several elements failed, and issues early alerts with recommended actions.

Sector

Energy and utilities: grid operators, distributors, large industrial consumers and regulators.

What it improves

Grid operation today is reactive and depends on expensive, slow simulation tools. Here the solver is proprietary and ten to fifty times faster, which allows anticipation instead of reaction.

Target client

Power grid operators and distributors, renewable asset management platforms and sector consultancies.

Status and maturity

Very solid base: proprietary power-flow solver, simulator with industry-standard datasets, trained decision agents, operations console and explicit alignment with the European energy digitalisation roadmap.

Differentiator

The power-flow solver is proprietary and free of commercial dependencies: no third-party licences and unrestricted optimisation.

Commercial fit

Multi-year contracts with operators and projects funded by European energy digitalisation programmes. Long sales cycle, high contract value, little competition.

Architecture

Layered architecture

A continuous operations loop over a digital twin of the grid, fed by governed data and solved by a proprietary power-flow engine. Without that engine there would be no anticipation: speed is what turns operation from reactive into anticipatory.

  1. 01

    Power-flow engine

    Proprietary AC and DC power-flow solver with a hot-compiled core. No commercial dependencies: the conventional tool is kept only as a validation reference, not as a dependency.

  2. 02

    Digital twin and forecasting

    A replica of the grid with a feature store, forecasting engine and climate scoring, where scenarios are rehearsed before touching the real network.

  3. 03

    Operations loop

    Sense, think, decide, act, verify and learn, with a supervisor above. Early warning, a matcher for patterns that preceded real incidents, and adversarial scenarios.

  4. 04

    Decision agents

    An independent simulator where reinforcement-learning agents are trained — including tree search and multi-agent coordination — on industry-standard datasets.

  5. 05

    European regulatory package

    Modules aligned with the European roadmap for digitalisation and AI in energy: renewables, flexibility and local market, operation, data centres as flexible load, explainability, model registry, federated learning, AI-regulation compliance, grid-model interoperability and network cybersecurity.

  6. 06

    Governed data

    Every external source — weather, generation, prices, hydrology, hazards — enters through an auditable layer with a per-environment key, no personal data, and a degraded mode if the provider fails.

Inventory

Status per component

A real inventory of the asset’s components with their declared status. We publish capability, never code or internal figures.

Built and testedFunctional with declared debtDesigned, not implementedWired; depends on a third party

  • Proprietary power-flow solver

    Built and tested

    Core of the asset and the origin of the declared speed advantage. Zero commercial dependencies.

  • Operations loop and digital twin

    Built and tested

    Full cycle with supervisor, emergency stop and human-in-the-loop above autonomy.

  • Simulator and trained agents

    Built and tested

    Environments, agents and configurations over industry-standard grid cases.

  • European regulatory package

    Built and tested

    Self-contained modules registered in the core and exposed through a proprietary interface, with a green test suite.

  • European package extensions

    Designed, not implemented

    Real topological reconfiguration, distributed-generation connectors, data spaces and formal adversarial robustness: declared pending, not sold as done.

  • External data integration

    Wired; depends on a third party

    Layer wired and audited; depends on the data provider, with an explicit degraded mode when unavailable.

  • Operations console

    Built and tested

    Proprietary interface for the grid operator.

  • Sovereign runtime

    Built and tested

    Proprietary supervisor and packaging: cloud, on-premise or air-gapped deployment.

This profile describes capabilities and status, not implementation. Code, architecture documentation and internal figures are shared under a confidentiality agreement during due diligence.

Platforms

Moeris

Cybersecurity

Prepares an organisation for the day today’s cryptography stops protecting its data.

Solid base Read the profile

Hecate

Cybersecurity

Protects dozens of client companies from a single console, without mixing their data.

Solid base Read the profile

Nemea

Real estate and banking

Turns bank debt and real-estate portfolios into exploitable investment intelligence.

Solid base Read the profile

Infinity

Real estate and banking

Publishes new-build real-estate developments automatically, with its own viewer and the client’s brand.

Solid base Read the profile
Describe a case See all platforms