(H/F) AI & Control Senior Engineer – Manipulation

Wandercraft; Paris; Sur place; EUR 90 000 - 130 000; Plein temps

Résumé du poste

Wandercraft recherche un ingénieur sénior en manipulation robotique pour faire progresser les méthodes, du contrôle basé sur le modèle à l’imitation et diffusion, vers des capacités robustes sur nos robots humanoïdes Calvin.

Le rôle mêle recherche, expérimentation pratique et intégration, avec un accent fort sur la résolution de problématiques de manipulation en conditions réelles et sur le travail avec l’équipe AI & Control.

Qualifications

• Diplôme d’ingénieur ou Master requis, doctorat apprécié.
• Au moins 5 années d’expérience en robotique ou apprentissage automatique, manipulation incluse.
• Expérience académique/chercheur en robotique, contrôle optimal, imitation learning ou diffusion.
• Expérience en gestion de jeux de données sur le long terme, stockage et CI.
• Esprit créatif et capacité à prototyper rapidement sans infra existante.
• Projets robotiques personnels ou professionnels valorisés.
• Maîtrise du développement logiciel collaboratif, modularité et CI.

Responsabilités

• Conduire des revues de littérature en manipulation robotique.
• Présenter les avancées et algorithmes développés à l’équipe.
• Développer des algorithmes de manipulation et entraîner des politiques.
• Tester les algorithmes sur robots réels et en simulation.
• Étudier les cas d’utilisation et proposer des solutions techniques.
• Améliorer le réalisme des simulateurs et comparer environnements.
• Développer tests automatisés, améliorer qualité du code et pipelines de vérification.
• Participer à la conception produit en expliquant les exigences hardware (vitesse, capteurs, calculs).
• Démontrer des tâches et collecter des démonstrations pour apprentissage.
• Voyager chez les clients pour valider les cas en conditions réelles.

Connaissances

Python

C++

Machine learning

Imitation learning

diffusion models

Robotics & control systems

Linear algebra

Numerical optimization

Computer vision

Software engineering best practices

Collaborative software design

Continuous integration

Automated testing

Formation

Engineering degree or MSc

PhD appreciated

Description du poste

Version française de l’offre disponible sous demande.

Wandercraft is strengthening its Manipulation team to tackle challenging industrial use cases with Calvin, our humanoid robot.

We are looking for a senior engineer with strong expertise in robotic manipulation to help us push the state of the art and turn advanced approaches — from model-based control to imitation learning and diffusion policies — into robust capabilities on real robots.

The role combines research, hands‑on experimentation, and integration, with a strong focus on solving real‑world manipulation problems.

Reporting to the Head of Manipulataion within the AI & Control department, the AI & Control Senior Engineer will be in charge of (non‑exhaustive list):

• Conducting literature reviews, particularly in robotic manipulation

• Presenting the latest research advances and internally developed algorithms to the team

• Developing manipulation algorithms and training policies

• Testing algorithms on real robots and in simulation

• Studying use‑cases and proposing technical solutions

• Improving simulator realism and comparing different simulation environments

• Developing automated tests, improving code quality, and designing verification pipelines to prevent regressions

• Participating to the product design by explaining hardware requirements related to manipulation (required speeds, sensing, computing, …)

• Demonstrating tasks and collecting demonstrations for learning experiments

• Traveling to customer factories to validate use cases in real‑world conditions and troubleshoot issues when necessary

• Diploma : Engineering degree or equivalent Master’s degree required, PhD in a relevant field is very appreciated

• Required experience: At least 5 years of experience in robotics or machine learning, including experience in robotic manipulation. Academic or research experience is highly valued, particularly in robotics, optimal control, imitation learning, or diffusion models. Ideally, you have experience managing datasets over time, including dataset storage, corrections, augmentation, usability, tooling, and continuous integration. You are creative and innovative, and comfortable in developing new solutions with little existing infrastructure and rapidly building prototypes. Personal or professional robotics projects are highly valued. You enjoy experimental work and adapting algorithms to the constraints of real‑world environments and commercial products. You have experience with collaborative software development practices, including modular design, test‑driven development, and continuous integration. You are comfortable working with linear algebra and matrix computations, kinematics, and physical modeling, including coordinate transformations and velocity computations. You also have strong software development experience in Python and C++. You are rigorous, able to review your own work, and can clearly explain your calculations and algorithms. You are curious and analytical, able to stay up to date with new technologies and learn effectively from scientific literature and interactions with the research community at conferences and other events.

• Required skills: Python - Expert level C++ - Autonomous level Machine learning - Expert level Imitation learning and diffusion models - Trainer level Robotics and control systems - Expert level Linear algebra, matrices computations, numerical optimization - Autonomous level Computer vision - Autonomous level Software engineering best practices, collaborative software design, continuous integration, automated testing - Autonomous level