
Computer vision and deep learning in the real world
I lead the ML team at Onfido. We use computer vision and deep learning for online identity verification. From 2015 to 2021, I led the machine learning team for recommendation at Criteo. From 2010 to 2014, I led the computer vision team at Thales Optronics in France.
I graduated with a PhD at MIT in 2010 under the supervision of Prof. Seth Teller. I was a core member of the MIT team at the DARPA Urban Challenge in 2007. My work was published at CVPR, ICCV, ICRA and IJFR. I teach the deep learning class at ENSAE Paris.
- Dec 2025 · nano-trm is out!
- Sept 2025 · Identity in the Age of Deepfakes
- Sept 2024 · The Handbook of Applied AI teams
- Jan 2023 · AI has crossed the Rubicon. Now what?
- Dec 2022 · Building a recommendation engine from scratch
- Sept 2022 · Building without Bias at Onfido
- Aug 2022 · Where self-driving is headed
- Jan 2022 · Lightweight Representation Learning For Efficient And Scalable Recommendation
- Dec 2021 · The high road of performance review
- May 2020 · The tradeoffs of large-scale machine learning: the price of time
- Feb 2020 · Your first 90 days as a tech lead
- June 2019 · Making your company machine learning-centric
Also on Medium
Featured talks
- Oct 2024 · Fraud Prevention with Computer Vision in the GenAI age, AIAI Boston · slides
- June 2023 · Bringing automation and fairness to identity verification, ML Prague · slides
- June 2023 · An introduction to ChatGPT and generative AI, Point Nine CTO Meetup · slides
- March 2023 · ChatGPT: have we crossed the Rubicon? BPCE DigiPay Tech+ Event
- 2022 · IFTTD podcast: Large-scale AI (in French) · youtube
- 2020 · Job interview talk at Onfido, Scalable representation learning
- 2018 · 7th RecSys London Meetup, Recommending Products From a Retailer That the User Never Visited Before
- 2017 · Criteo RecSys London, Product recommendation beyond collaborative filtering
- 2016 · ICML Online Advertising Systems Workshop
- Oct 2014 · ICIP Workshop, Image and Video Processing for Defense and Transportation
Publications
- Lightweight representation learning for efficient and scalable recommendation, arXiv:2101.00870v2, Nov 2021
- Ground robot navigation using uncalibrated cameras, ICRA 2010
- PhD Thesis, 2010
- Body-relative navigation guidance using uncalibrated cameras, ICCV 2009 · slides
- A perception-driven autonomous urban vehicle, IJFR 2008
- Wide-area egomotion estimation from known 3d structure, CVPR 2007
- SLYK: A Transparent Fault-Tolerant Migration Platform, MIT Technical Report, 2005
- The City Scanning Project, MIT Technical Report, 2002
Also on Google Scholar
Teaching
- Deep Learning class at Master level at ENSAE, since Spring 2024
- M2 level at Paris Dauphine University, IASD, 2019-2020
- Workshop on Recommendation, Ecole Polytechnique Data Science Summer School, June 2018 & 2019
- Java at MIT over IAP, 2008 – one of the first courses on OpenCourseWare
Misc
- Book recommendations for tech leaders
- Co-founded Reveal, the RecSys workshop for causal recommendation with Microsoft, Netflix and Cornell. Now continues as Consequences.
- Won the Criteo 2015 Hackathon (16 teams) with image-based similarity using deep networks
- Google Hashcode: 5/65 in 2015 Finale, 76/1054 in 2016, 300/2815 in 2017
- Advent of Code 2023 · blog post · github