Technology transparency

What powers Sympheny — and what doesn't.

A clear explanation of how Sympheny calculates, what the clustering does and does not do, and where AI assistants fit — plus precisely what happens to your data.

Urban Sympheny AG
Sympheny Platform
Audience
Customers, partners & procurement
Version
2026
How Sympheny works

Three distinct layers — each transparent and purposeful.

Sympheny operates across three layers. The first two run inside the calculation: the optimiser itself, and the clustering that makes a full year of hourly data solvable. The third sits outside it — the interfaces you can drive Sympheny through, including AI assistants you choose to connect. Nothing in the third layer changes what the first two compute.

Layer 1

MILP mathematical optimisation

Always active

The core of Sympheny is a Mixed-Integer Linear Programming (MILP) engine — a rigorous mathematical approach used in engineering and operations research. It finds the optimal energy system configuration given your technical, economic, and regulatory constraints.

Every result is fully deterministic and auditable: the same inputs always produce the same outputs, and every recommendation can be traced back to its underlying equations. There is no black box, and no AI in the calculations.

Deterministic Fully auditable Physics-based No AI in calculations
Layer 2

Clustered profiles

Always active — you control the resolution

Solving a full year of hourly data directly would be impractical, so Sympheny groups similar days into a smaller set of representative typical days using unsupervised clustering. You control how many through the temporal resolution setting, which also fixes the error tolerances the clustering has to meet.

This is grouping, not learning. No model is trained, nothing is carried between runs, and your data is never used to improve anything. The clustering is computed fresh from your own inputs every time, and re-running an unchanged scenario produces the same typical days and the same results.

Unsupervised clustering No model trained, nothing retained Reproducible run to run You control the resolution
Layer 3

AI assistants and agents

Off by default — you connect them

Sympheny publishes a REST API, a Python SDK and an MCP server, so an AI assistant or coding agent can set a scenario up, start an optimisation run and read the results back. Setup is documented at docs.sympheny.com. This is an interface to the platform, not a change to how it calculates: an assistant can build the model, but the model is still solved by the MILP engine in Layer 1, and the same inputs still produce the same outputs.

None of this is on by default. You decide whether to connect an assistant and which provider to use, and the data you send reaches that provider under your agreement with them, not ours. Sympheny embeds no third-party model in its own computation pipeline.

Off by default — you connect it Builds models, never solves them REST API · Python SDK · MCP Your choice of provider
Our data commitments

Four promises we make to every customer.

Your data is never used for training

No customer operational data — energy profiles, building data, or project inputs — is ever used to train any model, ours or third-party.

Full result auditability

Every optimisation output can be traced to its mathematical inputs and constraints. We can explain every recommendation.

You choose what connects to Sympheny

AI assistants and agents are optional and customer-connected. Sympheny ships no embedded assistant and connects nothing on your behalf.

Built in-house, not outsourced

All algorithms and models are developed by the Sympheny team. We do not embed third-party LLMs or AI services into the computation pipeline.

"

Sympheny is built on a mathematically rigorous MILP core, supported by unsupervised clustering that groups your data without ever learning from it — reproducible run to run, and with no model trained on customer data.

Customer FAQ

Questions about AI & data in Sympheny.

Straightforward answers to the questions we hear most often from customers, partners, and procurement teams.

Data & privacy

Is my operational data used to train Sympheny's models?

No. Customer data — energy profiles, load curves, building inputs, or any project-specific information — is never used to train any model. There is no trained model in Sympheny to begin with: the clustering is computed fresh from your inputs on every run and retains nothing.

Does Sympheny send my data to any external AI service or third-party model?

No. All computation, including the clustering, runs within the Sympheny platform using our own algorithms. No customer data is transmitted to external AI providers, large language models, or cloud-based inference services. If you choose to connect your own AI assistant to the Sympheny API, that is a separate arrangement between you and that provider.

Is the clustering machine learning?

It is unsupervised clustering, which is conventionally classed as machine learning — our product documentation describes the output as machine-learning-generated clustered profiles. The distinction that matters is that it groups data rather than learning from it. No model is trained, nothing is retained between runs, and the same scenario always produces the same clusters.

Note: If you require a formal data processing addendum or documentation for procurement purposes, Sympheny can provide this upon request.

Where is my data stored and processed?

Project data is hosted on Microsoft Azure (Switzerland and the EU) and AWS in Stockholm by default, with additional regional deployment options available for enterprise customers. Storage and processing details are governed by your specific service agreement — contact your account representative for specifics.

About the technology

Is Sympheny an AI tool?

No — Sympheny is a mathematical optimisation platform. Its core is a Mixed-Integer Linear Programming (MILP) engine, a well-established, deterministic approach used in engineering and operations research worldwide. Unsupervised clustering prepares your input data for the solver, and it learns nothing. Separately, you can connect your own AI assistant to drive Sympheny through its API — that changes how a model is built, not how it is solved.

What do you mean when you mention AI features?

Two things, both narrow. First, the clustering that groups similar days into typical days so a year of hourly data can be solved — unsupervised, no model trained, nothing retained. Second, AI assistants that you choose to connect from outside, which can set scenarios up but never perform the optimisation. Neither replaces the mathematical optimisation.

Can I explain the results to my client or regulator?

Yes. Because the core optimisation is MILP-based, every result is fully traceable: you can point to the specific constraints, cost parameters, and energy balances that led to a given recommendation. Sympheny does not produce outputs that cannot be explained from first principles.

Who built the algorithms — is this based on a third-party AI platform?

All algorithms and models are developed and maintained by Sympheny's engineering team. We do not use embedded third-party AI services (such as GPT, Gemini, or similar large language models) within the computation pipeline.

Working with AI assistants

Can I use Sympheny with an AI assistant or agent?

Yes. Sympheny publishes a REST API, a Python SDK and an MCP server, so an assistant or coding agent can create scenarios, start optimisation runs and read results back. Setup is documented at docs.sympheny.com. Nothing is connected by default — you decide whether to use it.

If I connect an assistant, where does my data go?

To whichever provider you connect. If you point an assistant at Sympheny, the scenario data it sends and receives passes through that provider under your agreement with them, not ours. Sympheny does not choose the provider for you and embeds no third-party model in its own computation pipeline.

Does connecting an assistant change my results?

No. An assistant can build or modify a scenario, but the optimisation always runs on the same deterministic MILP engine. A scenario assembled by an assistant and the identical scenario assembled by hand produce the same results.

Note: In short: an assistant can set the model up. It never solves it.

Procurement & compliance

Can Sympheny provide documentation for our AI governance or procurement process?

Yes. Sympheny can provide a technical description of the platform's AI components, a data processing summary, and — where required — a formal Data Processing Agreement (DPA). Please contact your account representative to initiate this.

Is Sympheny compliant with the EU AI Act?

Sympheny monitors regulatory developments including the EU AI Act and is committed to maintaining compliance as requirements come into force. Sympheny's core is a deterministic mathematical model with no trained AI component, which places it well within transparency and auditability requirements. Where a customer connects their own AI assistant, that assistant is the customer's system and falls under their own governance. For formal compliance documentation, please contact us directly.

Who can I contact with further questions?

For technical, legal, or procurement questions about data handling and AI usage in Sympheny, please reach out to your Sympheny account representative or contact us at contact@sympheny.com.

Need procurement documentation?

We can provide a Data Processing Agreement, sub-processor list, and a technical description of the platform's AI components on request. Talk to us before procurement — we make the review easy.

Chat on WhatsApp