Stéphane Schuck
Founder · Physician · Public Health Expert
Clinical and public health leadership anchoring Ultima-i in medical seriousness, epidemiology and health-system relevance.
Understand.
Generate.
Predict.
Ultima-i makes complex
health data clearer,
more useful and
easier to act on.
About Us
Scientific depth, product discipline and a clear health mission.
Ultima-i sits at the intersection of medicine, public health, data science and applied artificial intelligence. The company is designed to help organisations move from complexity to usable health intelligence.
4core team members across medicine, public health, pharma, finance and AI engineering.
20+years of experience represented across epidemiology, study design and scientific delivery.
35+years of operational data and infrastructure expertise represented within the leadership team.
300+scientific publications across medicine, epidemiology and AI.
Why Ultima
AI is stronger when it stays anchored in medicine.
Ultima-i brings together physicians, public health experts, pharma specialists, data infrastructure leaders and AI engineers to solve health questions with more precision. That combination helps teams move faster without losing scientific discipline, readability or practical relevance.
What clients gain. A partner able to frame a problem medically, structure it scientifically and turn it into a tool, workflow or platform that people can actually use.
Why it matters
From complexity to action.
Health organisations do not need abstract AI promises. They need tools and programmes that make complex data easier to understand, generate and operationalise.
Our Vision
Health AI that earns trust.
We believe better systems should improve decision quality without becoming opaque. The objective is not more technical noise, but clearer evidence, better products and more usable intelligence.
Our Story
Built from complementary expertise.
Ultima-i was formed around a simple idea: health AI becomes more valuable when medicine, public health, rigorous methods and product execution are designed together from the start.
Partners
Trusted collaborations across research, care and health innovation.
Ultima-i works alongside hospital, research, industry and innovation partners on programmes where medical depth, operational clarity and AI execution need to move together.
Approach
Medical, technical and operational.
Each engagement is framed through domain understanding, scientific method and delivery discipline so the outcome remains useful beyond the prototype stage.
Team
The people behind Ultima-i.
A multidisciplinary team combining medical evaluation, public health, scientific operations, data strategy and generative AI execution.
Charles Huot
Pharma Strategy · Industry Expertise
Brings operational understanding of pharmaceutical environments, programme framing and strategic execution in health contexts.
Bertrand Delezoide
AI Expert · Data Science
Leads advanced modelling, AI structuring and technical translation between scientific ambition and usable systems.
Amine Chouaki
AI Engineer
Develops generative and predictive AI workflows, with a focus on health data usability, structured outputs and product implementation.
Partner network
Research, care, industry and innovation partners.
Our collaborations span academic research, hospital ecosystems, pharmaceutical programmes, data science and innovation funding. Each partnership reinforces the same objective: making health data more actionable.
Ultima Platform
A dedicated platform page is prepared for the next stage.
This section is intentionally kept light for now. The full platform experience will be developed as a separate, deeper product showcase.
What this page will become
The platform page is intended to present the Ultima environment as a clear, premium and highly functional health AI workspace. It will be developed as its own major piece rather than condensed into a quick placeholder.
Product storytelling
Sharper explanation of modules, logic and use cases.
Interface preview
Dedicated screenshots or a live-feel presentation layer.
Governance and workflow
A clearer view of how the platform supports real teams.
Understand
Make dense health information readable.
Interpret clinical language, longitudinal signals and large-scale health data with medical context, evidence and operational clarity.
Generate
Create assets that move science forward.
Generate synthetic data, structured outputs and new research material under medical and scientific constraints.
Predict
Build prediction around real patient questions.
Design interpretable predictive systems for trajectories, next events and decision support where timing and action matter.
Understand
Interpret complex health information with medical context, evidence and clarity.
- Clinical and regulatory language workflows
- Signal review, evidence synthesis and data interpretation
- Patient journey and longitudinal analysis
Generate
Create controlled data and outputs that enable rigorous experimentation, model development and validation.
- Synthetic data generation
- LLM-assisted workflows and structured outputs
- Content and data generation under scientific constraints
Predict
Build predictive systems that remain interpretable, operational and connected to real health questions.
- Next-event and trajectory prediction
- Risk-oriented modelling and decision support
- Health AI systems designed for operational use
Projects
Concrete projects, each built around a clear problem.
The examples below show the type of work Ultima-i is building today across health data, generation and prediction.
Project perspective
From population-scale complexity to usable decisions.
Ultima-i projects often start with dense clinical or real-world complexity, then move toward tools, models and workflows people can act on.
Patient-level view
Keep the individual visible.
Even when the work is large scale, the final objective remains concrete: clearer reasoning, better timing and more useful decisions for real patients.
Project 01
SIDARTA
A safety intelligence environment designed to help teams review information faster, prioritise the right signals and coordinate action more clearly.
SIDARTA is developed in collaboration with a major pharmaceutical industry player. It is designed as a workspace for pharmacovigilance and safety teams facing large volumes of information, multiple viewpoints and constant pressure for consistency.
Need
Make safety review less fragmented across people, risks, patient profiles and decision timelines.
Response
Interactive risk views, patient-level reasoning, key risk analysis, probability trees and embedded AI assistance.
Value
Better visibility, clearer prioritisation and a stronger bridge between analysis and action.
Project 02
Silico
An end-to-end programme for patients with ciliopathies, combining understanding, generation and prediction in one medical workflow.
Silico is developed in collaboration with Institut Imagine and AP-HP. It is designed to support a demanding clinical objective: better understand the disease, generate the right data assets and identify the most relevant timing to administer a treatment.
Need
Clarify complex disease mechanisms and patient patterns in a context where timing matters clinically.
Response
Combined disease understanding, targeted data generation and predictive modelling around treatment timing.
Value
A more informed path toward clinically useful decisions for patients affected by ciliopathies.
Project 03
SYNTHIA
A synthetic health data initiative designed to generate credible, useful and privacy-conscious datasets for experimentation and development.
SYNTHIA focuses on the practical value of synthetic data in health contexts where access, privacy or scarcity can limit exploration. The goal is not artificial volume for its own sake, but data that can support better testing, better models and better scientific workflows.
Need
Access realistic health data assets without exposing sensitive information or waiting for impossible collection conditions.
Response
Controlled synthetic data generation with strong attention to quality, utility and downstream evaluation.
Value
Faster experimentation, safer exploration and more room to design and test health AI systems.
Project 04
Longitudinal Data
A project focused on patient trajectories, temporal structure and the interpretation of complex longitudinal health data.
This work addresses one of the hardest realities in health data: value often sits in the sequence, not only in isolated variables. Ultima-i uses that longitudinal logic to make trajectories more readable, comparable and actionable.
Need
Understand how events unfold over time rather than analysing health data as disconnected snapshots.
Response
Temporal structuring, trajectory analysis and modelling approaches aligned with real patient histories.
Value
More faithful interpretation of patient journeys and stronger foundations for downstream prediction.
Project 05
Next Event Prediction via Fine-Tuning
A predictive modelling effort designed to anticipate the next clinically relevant event from patient history and structured context.
This project explores how fine-tuned models can capture sequential logic in health data and turn it into useful predictive support. It is particularly relevant when the next event matters for care planning, follow-up or operational decisions.
Need
Anticipate what is most likely to happen next in a sequence of clinical or patient events.
Response
Fine-tuned predictive models built around temporal health data and clinically meaningful event definitions.
Value
More proactive workflows, clearer anticipation and stronger decision support across complex patient pathways.
Contact
Let’s discuss a health, data or AI problem worth solving.
Ultima-i is best suited to conversations where medical seriousness, technical depth and strategic clarity all matter at once.
What to discussSafety intelligence, clinical NLP, generative AI, synthetic data, evidence programmes and predictive health products.
Who it is forPharma, biotech, hospitals, scientific teams, public health organisations and innovation partners.
Emailcontact@ultimai.ai
LocationParis, France