Job Offer Physora - ML Application Engineer

Transvalor Sa; Biot; Hybrid; EUR 45,000 - 60,000; Full time

Benefits offered by this job

Housing assistance

Daily transport support

RTT (reduction of working time)

Job summary

Physora - A Transvalor Company is seeking an ML Application Engineer to deploy surrogate models in real-world manufacturing settings. You will assemble technology components, run models at customer sites, and build intuitive interfaces for operators.

You will collaborate with R&D to adapt architectures and ensure reliable performance, including on-site commissioning and production monitoring across the shop floor.

Qualifications

• Experience deploying ML models to production with at least one end-user system.
• Daily use of PyTorch and scientific Python (NumPy, Pandas, SciPy).
• Experience with industrial time-series data and sensor data.
• Strong software engineering discipline: Git, testing, readable code.
• Willing to spend weeks on site and talk to shop floor and quality engineers.
• Engineering degree or master's in a relevant field.

Responsibilities

• Prepare training data by process and customer: simulation outputs and shop-floor signals.
• Train, calibrate and validate surrogate models for customer use cases.
• Build end-to-end inference pipeline with plant latency in mind.
• Develop interfaces and tooling for operators to monitor predictions.
• Participate in on-site commissioning and production monitoring.
• Turn field failures into reproducible R&D data and insights.
• Contribute to productization: campaigns, connectors, dashboards.

Skills

ML in production

PyTorch

Scientific Python (NumPy, Pandas, SciP

Education

Engineering degree or master's in CS/Math/ME/Industrial Eng

Tools

Python (NumPy/SciPy/Pandas)

OPC-UA

MQTT

AMQP

Git

Job description

Job Offer Physora - ML Application Engineer

Transvalor is hiring an ML Application Engineer (M/F) for its new subsidiary Physora

Job Title: ML Application Engineer
Company: Physora - A Transvalor Company
Location: Sophia Antipolis
Contract Type: Permanent (CDI)
Status: Executive

Reporting to: Industrial Project Manager

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 are the person who assembles the technology components to create an application that will run the surrogate model at the customer’s site.
The R&D team designs the architectures and demonstrates that they converge; you take them as they are and bring them to a real-world process, with signals, noisy presses, and geometries that change from one part reference to another.
You will also create the intuitive interface that will make the tool easy for operators to use.

MISSIONS

In this role, you will be responsible for:

• Prepare training data by process and by customer: outputs from simulation campaigns (FORGE®, THERCAST®) and shop-floor signals — matching, cleaning, quality control, versioning.
• Train, calibrate and validate surrogate models on the customer use case using the architectures delivered by R&D; document what was tried and what was retained.
• Build the end-to-end inference pipeline: signal acquisition via OPC-UA or message broker, preprocessing, model execution within the plant’s latency budget, and delivery of results to the user.
• Implement the acceptance metric defined with the Project Manager and R&D, measure it continuously and make deviations visible — including when they are unfavorable.
• Participate in on-site commissioning and production monitoring: signal drift, prediction drift, retraining.
• Turn field failures into reproducible cases for the R&D team — dataset, configuration, observed deviation. This is the main channel through which R&D learns what its models cannot do.
• Contribute to the tooling that will turn this first project into a product: campaign scripts, connectors, result dashboards.
• Develop the interface that will allow users to monitor the model’s predictions.

Positioning

• Model architecture belongs to R&D. You apply it, and you challenge it with data, rather than rewriting it: a variant that you consider necessary is a request addressed to R&D, not a local branch.
• Delivery responsibility and the customer relationship belong to the Industrial Project Manager. You are in direct contact with the shop floor and the user on technical matters; scope and schedule commitments go through them.

PROFILE

• Experience putting ML models into production — at least one system that has actually run for an end user, with all the problems that come with it. Outside an academic context.
• PyTorch and scientific Python (numpy, pandas, scipy) on a daily basis; comfortable reading, modifying and debugging training code written by others.
• Experience with industrial time-series data: timestamps, heterogeneous sampling frequencies, missing values, sensors that lie.
• Strong software engineering discipline: Git, testing, code that others can read, reproducible results.
• Enjoys being on the ground. You are willing to spend weeks on site, and you know how to talk to a machine setter as well as to a quality engineer.
• Engineering degree or master’s degree in computer science, applied mathematics, mechanical engineering or industrial engineering.

nice to have

• Knowledge of forming processes (forging, casting), presses and their instrumentation.
• Finite element methods and numerical simulation; experience with a solver (FORGE®, THERCAST® or equivalent).
• Industrial protocols (OPC-UA, MQTT, AMQP); edge deployment on constrained hardware, containers.
• GNNs or neural operators — useful for communicating with R&D, not required.

POSITION DETAILS

• Location: Biot / Sophia Antipolis;
• Full-time permanent contract (CDI);
• Occasional work from home;
• Annual gross salary: between €45 and €60k;
• 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