Software engineer on private and decentralized machine learning (H/F)

Inria, the French national research institute for the digital sciences; France; Hybride; EUR 45 000 - 70 000; Plein temps

Résumé du poste

A leading research institute in France is looking for an engineer to contribute to a collaborative project on Federated Learning. The role involves extending a machine learning library, conducting experiments, and publishing scientific results. Strong Python programming skills and good communication abilities are required. The position offers flexible working hours, the possibility of teleworking, and a supportive work environment.

Prestations

Partial reimbursement of public transport costs

7 weeks of annual leave plus additional days off

Possibility of teleworking

Professional equipment available

Social and cultural events

Qualifications

• Experience with object-oriented programming, unit testing, and documentation writing.
• Interest in machine learning and medical applications.
• Familiarity with git workflow and community software development.

Responsabilités

• Consolidate and extend the existing library for decentralized and privacy-preserving machine learning.
• Publish open source code and integrate into existing libraries.
• Conduct experiments with medical partners on multicentric medical studies.

Connaissances

Programming skills in Python

Good understanding of scientific papers on machine learning

Good communication skills

Description du poste

Offer Description

The position will be supported by FedMalin, a collaborative project on Federated Learning between 11 teams at INRIA. The project addresses FL challenges when deployed over the internet (privacy, heterogeneity, energy, fairness, ...) and has medicine as a main targeted application domain.

FedMalin develops several software tools, including the open source library DecLearn (https://gitlab.inria.fr/magnet/declearn/declearn2) for private and decentralized/federated machine learning and data analysis. The hired engineer will contribute to the ongoing development of DecLearn, expanding its capabilities with new algorithms and enhanced functionalities.

The activities will include interactions with the members of the project, the Magnet and Premedical teams (researchers and engineers). We also expect to conduct multi-centric medical studies across several hospitals. The activities can also include travel, e.g., to conferences to demonstrate the developed library and to contribute to the community building effort.

• Consolidate and extend the existing library for decentralized and privacy-preserving machine learning developed in the project
• Deploy the library in real-world conditions and experiment on synthetic and (benchmark) medical data, analyzing the benefits and the costs compared to a centralized approach.
• Publish open source code and integrate in existing libraries
• Publish scientific results in medical and computer science conferences
• Implement federated and privacy-preserving algorithms for machine learning
• Experiment with medical partners onmulticentric medical studies
• Evaluation of results
• Reporting, disseminating and presenting results
• Programming skills in Python, including object oriented programming, unit testing, documentation writing, deployment tools, asynchronous programming and networking.
• Good understanding of scientific papers on machine learning.
• Interest for machine learning and medical applications.
• Good communication skills; communication and animation of software development communities, git workflow

Languages

• FRENCH – Basic
• ENGLISH – Good

Additional Information

• Partial reimbursement of public transport costs
• Leave: 7 weeks of annual leave + 10 extra days off due to RTT (statutory reduction in working hours) + possibility of exceptional leave (sick children, moving home, etc.)
• Possibility of teleworking and flexible organization of working hours
• Professional equipment available (videoconferencing, loan of computer equipment, etc.)
• Social, cultural and sports events and activities