At althiqa, we are on a journey to make AI evaluation simple and trustworthy.Deeptech company (BFTE BPI).Incubated at Agoranov. SaaS: https://app.althiqa.io/Medium: https://medium.com/@althiqaResponsibilities:- Development of the SaaS platform- Scientific monitoring of latest research in AI and trustworthy AI- Producing our own research (2 papers accepted at NeurIPS 2022, 4 ongoing works)- Management of a team of developers and UX/UI - Communication on social media / seminars / conferences- Business development

---- Research in Machine Learning and Statistics ----Topics : Fairness, Robustness, Explainability, Correction of Bias, Transfer Learning, Sampling, Selection, Concentration.Several publications:- NeurIPS 2022 - What are the best systems? New perspectives on NLP Benchmarking. - Beyond Mahalanobis-Based Scores for Textual OOD Detection- ICML 2022 - Mitigating Gender Bias in Face Recognition using the von Mises-Fisher Mixture Model- ACL 2022 - Learning Disentangled Textual Representations via Statistical Measures of Similarity- NeurIPS 2021 - Online Matching in Sparse Random Graphs: Non-Asymptotic Performances of Greedy Algorithm- ICML 2021 - Learning from Biased Data: A Semi-Parametric Approach---- Teaching activities for the post-masters degrees Big Data and Artificial Intelligence ----- Practical sessions for the Machine Learning course- Data challenge organization- Supervision of internships (Computer Vision at Sicara, Causal Machine Learning at Air Liquide, Active Learning at Safran)

Applied mathematics tutorials for students in psychology and economics. Courses taught: introduction to analysis, optimization, statistics.

With Nathanaël Enriquez and Laurent Ménard on random matrices and random graphs. Topics: spectra of large random matrices, spectra of large random graphs, community detection in sparse random graphs.

With Shen Lin on the study of random walks on random graphs. Topics: random walk theory, random graph theory, ergodic theory.