Reference : UMR7161-OANBAL-003
Workplace : PALAISEAU
Date of publication : Monday, August 1, 2022
Type of Contract : FTC Technical / Administrative
Contract Period : 12 months
Expected date of employment : 1 November 2022
Proportion of work : Full time
Remuneration : Between 2109.46 and 2226.89 monthly gross depending on experience
Desired level of education : 5-year university degree
Experience required : Indifferent
The project ANR PROPEOS aims to propose innovative algorithms which preserve the advantages of personalization but give a core role to user privacy, algorithmic transparency, and trust by creating and exploiting user profiles exclusively on their devices.
The engineer help to conceptualize and analyze such system.
Create algorithms that can personalize services while ensuring users' privacy by exploiting the local profiles generated in T3; that provide meaningful explanations to both users and service providers about their outputs; and that does not discriminate or exploit users' vulnerabilities. While big companies such as Amazon have all the data necessary to provide relevant personalization of their products, smaller companies do not have the data and the expertise to do the same. Our platform is specifically dedicated to small companies and services to help them personalize services in a privacy-preserving way.
The engineer will focus on three tasks: Dual-source Personalization, Explainable Privacy-Preserving Personalization, Dependable Personalization.
- Master's degree in computer science or an engineering school.
- Strong python programming and system administration skills.
- Expertise in data analysis.
The work will take place at LIX, within the CEDAR research team. This laboratory is located on the campus of the Ecole Polytechnique in Palaiseau.
The work takes place within the framework of the ANR PROPEOS project.
The engineer will work on the direction of Oana Goga
The engineer will work in partnership with other researchers, doctoral students and other computer science research engineers.
Constraints and risks
We talk about it on Twitter!