Symptems

01 The founder

A doctor who helps
shape the standards —
and builds to them.

Eric Sassel, PhD sits on Luxembourg's EHDS medical-digitalisation committee, tasked with making the country's medicine coherently and interoperably digital by 2029, across the entire profession and the Greater Region.

He works on the committee defining how the country digitalises its medicine, and builds Symptems to those standards. Symptems is where that mandate becomes product: patient records that securely follow the patient wherever care is delivered. Continuity of care, real medical collaboration, a system that finally talks to itself.

And it's private by construction. Symptems runs on local, purpose-built AI: clinical data is processed on-site, not shipped off to a third-party model. GDPR isn't a checkbox bolted on at the end, it's the architecture, so patient data stays where it belongs and the practice stays in control.

“True care always listens. So we built the AI that lets doctors listen more, not decide less.”

Eric Sassel, PhD, interviewed on RTL Kloertext

Eric Sassel, PhD · Founder RTL Kloertext

02 The team

Physician, scientist, engineer. The right trio.

Dr Eric Sassel

CEO · Physician

Founder and practising physician. Brings the clinical reality and the national health-data digitalisation view straight into the product.

EHDS committeeRegulatory

Pr Michel Mittelbronn

CMO · Neuropathologist

Internationally recognised neuropathologist, co-author of the WHO classification of CNS tumours. Former PEARL / FNR.

FNRLNS

Nicolas Fritz

CTO · AI engineer

Has delivered projects for Fortune 500 firms and large-scale public-sector organisations.

Multi-agent systemsData security

03 What Symptems is

Three products. One nervous system for the practice.

For doctors

AI-augmented doctors

An AI at your side in consultation, handling the data, the documentation, the noise, so you get your time back to listen and to decide. Not to decide for you. Real medicine, not a demo.

In-consultDocumentationDecision support

For the whole practice

An AI-augmented practice

An assistant, for your assistants. Reception, back-office, follow-ups: the whole practice runs lighter. A quiet layer working behind every desk in the building.

Front deskBack officeAutomation

For patients

A patient-owned app

Patients chat with Symptems before the visit to bring context, and choose what to share, or not. GDPR end-to-end. No diagnosis, ever. Just deeper listening, and a far richer record for the doctor who receives them.

Patient-ownedGDPR e2ePre-visit context

The philosophy

LLMs hallucinate and cry wolf. A doctor doesn't need more noise. So we don't build an oracle. We build a copilot. Never an autopilot.

Aware of Europe's digital-health failures, and of exactly how brittle LLMs are, we engineered Symptems to help the doctor decide, never to decide instead. The human stays in the loop, by design, every single time.

04 The architecture

One AI agent per body system.

Not a fine-tuned monolith. A growing set of specialised agents, monitored into reliability, and built to swap in tomorrow's best model without a rewrite.

Neuro

Bone

Gastro

Immuno + more as we cover them

Agentic & multimodal, not RAG

The frontier moved months ago from one fine-tuned model to agentic, multimodal systems. We built for that frontier.

Model-agnostic by design

Any LLM plugs into any service. We snap in the most powerful model available, every time, as they ship, with no rebuild from scratch. Our architecture outlives any single model generation.

Probabilistic model, verifiable output

Multiple agents monitor the flow in real time, turning a probabilistic model into a dependable one. End-to-end checks at every step. Less hallucination. More trust where it actually matters.

Human in the loop, by design

The clinician stays the decision-maker. The AI proposes, never disposes, and every output is reviewable, traceable and overridable. Built to earn trust where it actually matters.

05 Regulatory readiness

Regulation as an engineered advantage, not a liability.

Our modular architecture separates the clinical decision-support module from the non-regulated modules. That decouples our go-to-market from the certification timeline: the modules outside medical-device scope generate usage and traction today, while the clinical module follows a documented conformity pathway.

Ships today · outside device scope

Non-regulated modules

Practice management Document sorting Scheduling Billing

Generating usage & traction now

MDR CLASS IIA/IIB · AI ACT PATHWAY

Clinical decision support

Symptom reasoning Per-function agents Clinical insight Patient app

Documented conformity pathway

Go-to-market, decoupled from certification

Modules that fall outside the medical-device scope generate revenue and adoption today. Certification gates the clinical module, not the company.

Boundaries drawn early, with counsel

We map the line between device and non-device modules from day one, and validate that classification with our regulatory advisors and, when the time comes, a notified body, rather than retrofitting it later.

Real-world evidence, the right way

We're fully aware the regulatory path is delicate and carries real legal weight. We're being advised across the board to build real-world clinical evidence under the appropriate framework, ahead of CE marking. None of this is legal advice.

06 Zero admin

Every document, sorted before you ask.

Biology, lab reports, faxes, mail, any source. A local, purpose-built AI extracts what matters, files it in the right patient folder automatically, and flags the urgent ones. The doctor decides what happens next. Local AI means GDPR by construction, not by paperwork.

A doctor who helpsshape the standards —and builds to them.