Job Offer Physora - ML Architect

Transvalor; Valbonne; Hybrid; EUR 55,000 - 70,000; Full time

Benefits offered by this job

Housing assistance

Transport allowance

RTT (time off)

Occasional work from home

Job summary

Physora, a Transvalor Group company, is seeking an ML Architect to design surrogate models for real-time production use. You will lead architecture direction, ensure models converge across geometries, and run in real-time within a customer-facing setting.

You will work under the Head of R&D, with opportunities to mentor PhD students and contribute to a cutting-edge AI for numerical material-forming simulations, based in Sophia Antipolis with occasional remote work.

Qualifications

• Geometric deep learning with meshes, point clouds or neural operators.
• Experience with Gaussian processes or Bayesian optimization.
• Expertise in PyTorch with multi-GPU training and debugging.
• Ability to validate against physics, not only data-driven results.
• PhD/Masters in ML, applied math, computational mechanics or physics.

Responsibilities

• Design architectures of surrogate models for production use.
• Train and debug surrogates on complex FE simulation data.
• Contribute to evaluation frameworks and model acceptance criteria.
• Mentor PhD students and collaborate with R&D leadership.

Skills

Geometric deep learning

Gaussian processes

PyTorch

Physics-informed validation

Multi-GPU training

Education

Master’s degree or PhD in ML, applied mathematics, computational mechanics or physics

Tools

Multi-GPU training

Job description

Job Offer Physora - ML Architect

Job Title: ML Architect
Company: Physora - A Transvalor Company
Location: Sophia Antipolis
Contract Type: Permanent (CDI)
Status: Executive
Reporting to: Head of R&D

COMPANY

Physora is a young Transvalor Group company, founded in 2026 and based in brand-new offices at the heart of the Sophia Antipolis technology park. It is dedicated to developing and industrializing AI surrogate models for the numerical simulation of material-forming processes, including forging, casting, welding and heat treatment.

These models can be integrated into the Group’s digital workflow—FORGE®, DIGIMU®, REM3D®, TRANSWELD®, THERCAST® and SIMHEAT®—to reduce cycle times from hours to seconds.

They will also enable us to expand into production and quality control, with real-time prediction solutions powered by high-quality simulations from the Group’s solvers.

POSITION

You design the architectures of the surrogate models that will be used in production by the application engineers.
The architectural direction is set by our Head of R&D; you are responsible for delivering it: models that actually converge, generalize from one geometry to another, and are designed to run in real time.
The nature of the role is first and foremost that of a technical expert working within an architecture you did not design, in a customer-facing company. If you are looking to make an ambitious architecture work on high-quality real industrial data, with more data than anyone else in this field can obtain, this role will be a good fit for you.

MISSIONS

• Implement, train and debug surrogate models on complex finite element simulation data.
• Address what makes the problem difficult in practice: variable mesh topology, remeshing events, multi-scale dynamics, distribution shift across geometries and process parameters.
• Contribute to the evaluation framework and to defining what constitutes an acceptable model.
• You embody the technical depth of the Physora thesis: you are its critical guardian, able to challenge the architecture whenever the data or results contradict it.
• Contribute to mentoring PhD students.

PROFILE

• Strong experience in one of the following fields (supported by publications, open-source contributions or production systems): Geometric deep learning: GNNs on meshes, point clouds, or neural operators.
• Strong experience in one of the following fields (supported by publications, open-source contributions or production systems): Gaussian processes, Bayesian optimization, kriging.
• Deep expertise in PyTorch. Multi-GPU training, profiling, and the debugging experience that comes from having seen models fail silently.
• A habit of validating against physics, not just against a test set.
• Master’s degree or PhD in ML, applied mathematics, computational mechanics or physics, or equivalent industrial experience.

YOUR STRENGTHS

• Finite element methods, computational mechanics or CFD.
• Diffusion models or flow matching, neural operators (DeepONet, FNO), learned physics simulators.
• Deploying models under real-world latency constraints.

POSITION DETAILS

• Location: Biot / Sophia Antipolis;
• Full-time permanent contract (CDI);
• Occasional work from home;
• Annual gross salary: between €55 and €70k;
• Executive status.

OUR BENEFITS

• An international, dynamic, and stimulating work environment;
• A positive atmosphere within teams passionate about digital technologies and scientific research;
• An attractive compensation and benefits package (aide au logement, prise en charge du transport quotidien, RTT, travail à domicile occasionnel).

More about Physora: www.physora.ai