Data Scientist Analyst
il y a 6 jours
Job Description:
About the job
Job purpose
Large Language Models (LLMs) have demonstrated impressive reasoning capabilities, but how do we reliably assess these abilities? Current benchmarks used to evaluate reasoning in AI models face important limitations regarding their robustness and their capacity to truly measure reasoning performance.
This internship is part of an ongoing PhD project that critically examines existing reasoning capabilities in LLMs. The research investigates:
- Benchmark robustness: How reliable are current evaluation methods?
- Assessment validity: Do these benchmarks actually measure reasoning abilities effectively?
- Model performance patterns: Which specific reasoning tasks do different models excel at, and how does this depend on their architecture, training approach, and size?
The ultimate goal is threefold: (1) gain deeper knowledge of LLMs' reasoning capabilities, (2) develop practical, user-friendly methods to assess model performance on specific reasoning tasks, and (3) contribute to improving models' reasoning abilities.
Main missions
Your responsibilities include:
- Benchmark Analysis: Evaluate the robustness and validity of existing reasoning benchmarks for LLMs
- Performance Characterization: Conduct systematic experiments to map which models perform well on specific reasoning tasks based on architecture and training characteristics
- Assessment Tool Development: Design user-friendly evaluation methods for specific reasoning tasks
- Improvement Strategies: Explore approaches to enhance reasoning capabilities in foundation models
Expected skills & experience
We are looking for someone with the following experience and skills:
- Familiarity with Machine Learning and Deep Learning techniques
- Solid understanding of natural language processing or large language models
- Familiarity with reasoning tasks (logical reasoning, mathematical reasoning, etc.)
- Knowledge of at least one deep learning library (PyTorch)
- Good Software Development skills (Python)
- Good English skills (French is a plus)
What we offer
We bring together the expertise, cultural diversity and creativity of over 8,000 employees worldwide and we're committed to equal opportunities in all aspects of employment (gender, LGBT+, disabled persons, or people of different origins) and to promoting Diversity & Inclusion by creating a work environment where all employees are treated with dignity and respect, and where individual differences are valued.
-
Data Scientist F/H
il y a 1 semaine
Paris, Île-de-France un emploi de Data Scientist FH Temps pleinData scientist F/H (Stage de 6 mois)Faire de la Data Science dans la musique, c'est analyser des millions d'écoutes quotidiennes, comprendre comment naissent les hits, comment évoluent les goûts et comment les artistes rencontrent leur public.Sony Music Entertainment France, un des leaders du secteur de la production, de la promotion et de la distribution...
-
Analyste Data scientist
il y a 5 jours
Paris, Île-de-France Choisir le Service Public Temps pleinInformations générales Organisme de rattachement Direction de l'administration pénitentiaire - Administration Centrale Référence Date de début de diffusion /01/2026 Date de parution /01/2026 Localisation PARIS Intitulé long de l'offre Ministère de la justiceDirection de l'administration pénitentiaireSous-direction de l'expertise...
-
cdi - consultant data scientist - h/f
il y a 2 semaines
Paris, Île-de-France Havas Data Business Intelligence Temps pleinDans le cadre de son hyper croissance, HAVAS DBi, l'agence conseil en data marketing du Groupe Havas, recherche unConsultantData scientist.2 ans d'expérience minimum dans le domaine de la data science et du data marketing.Rôle clé dans une agence en pleine croissance, avec une expertise sur des projets innovants. En étroite collaboration avec les...
-
Data Scientist
il y a 2 semaines
Paris, Île-de-France BlaBlaCar Temps pleinAbout BlaBlaCar BlaBlaCar is the world's leading community-based travel app enabling 27 million members a year to carpool or travel by bus in 21 countries. Our team of 800 employees counts over 50 nationalities and is spread across our 5 global offices, 30% working fully remotely. Your mission We're seeking a passionate Data Scientist to join our data...
-
Data Scientist
il y a 2 semaines
Paris, Île-de-France BlaBlaCar Temps pleinAbout BlaBlaCarBlaBlaCar is the world's leading community-based travel app enabling 27 million members a year to carpool or travel by bus in 21 countries. Our team of 800 employees counts over 50 nationalities and is spread across our 5 global offices, 30% working fully remotely.Your missionWe're seeking a passionate Data Scientist to join our data squads,...
-
Data Analyst
il y a 1 jour
Paris, Île-de-France Free-Work Temps pleinEn tant queData AnalystouData Scientist ou AI Ingineer, vous contribuez aux projets stratégiques de grands comptes bancaires en intervenant sur l'analyse, le traitement et la valorisation des données bancaires et financières.Vos principales missions :Recueillir et analyser les besoins métiersCollecter, structurer et traiter les données chiffrées issues...
-
Data Scientist
il y a 2 semaines
Paris, Île-de-France BlaBlaCar Temps pleinAbout BlaBlaCarBlaBlaCar is the world's leading community-based travel app enabling 27 million members a year to carpool or travel by bus in 21 countries. Our team of 800 employees counts over 50 nationalities and is spread across our 5 global offices, 30% working fully remotely.Your missionWe're seeking a passionate Data Scientist to join our data squads,...
-
Data Scientist H/F
il y a 7 jours
Paris, Île-de-France Data sea Temps pleinÀ propos du posteNous recherchons un(e) Data Scientist pour rejoindre notre équipe et participer activement à la valorisation de nos données. Votre rôle sera essentiel pour comprendre nos performances, anticiper les tendances et proposer des solutions basées sur la data.Vos responsabilitésExploiter et analyser de grands volumes de donnéesConstruire...
-
Data Analyst Senior – Data Management
il y a 5 jours
Paris, Île-de-France RIDCHA DATA Temps plein? ContexteDans le cadre d?un programme stratégique de gouvernance des données, une grande institution financière recherche un Data Analyst Senior pour renforcer son équipe Data & Innovation.La mission vise à améliorer la qualité, la conformité réglementaire et la maîtrise des données critiques utilisées dans les processus métier.?? Missions...
-
Data Analyst
il y a 2 semaines
Paris, Île-de-France Temeritati Temps pleinTemeritati est un cabinet de conseil en management spécialisé dans les secteurs financier et industriel. Notre expertise repose sur le triptyque « Risques & Conformité », « Efficacité opérationnelle » et « Transformation digitale »Dans le cadre de notre développement, nous recherchons des profils pour renforcer notre pôle d'expertise avec...