Andreas Roessler
Summary
🤩
RockstarAndreas Roessler is a CV/ML software engineer with 11 years of experience specializing in large-scale generative image and video models, currently working on diffusion-based avatar research at Meta Reality Labs. He spent six years as a research associate at TUM where he created the FaceForensics++ dataset and published influential work on facial manipulation detection presented to the German government. Andreas has led end-to-end development of production-grade synthetic avatar and multilingual dubbing systems, translating research prototypes into scalable deployments. He combines deep academic rigor—a mathematics master’s with top honors—with hands-on engineering across data strategy, model research, and productionization. Known for organizing community-facing workshops and courses, he bridges media forensics, generative modeling, and practical ML engineering. A detail often overlooked: his background in algebraic curves and computational intelligence informs a mathematically grounded approach to diversity and conditioning in generative models.
11 years of coding experience
Github Skills (4)
face-alignment10
face-detection10
face10
deepfake10
Programming languages (1)
Github contributions (5)
ondyari/FaceForensicsDataProcessing
Apr 2018 - Jan 2020
Contributions:15 commits in 1 year 9 months
ondyari/FaceForensics
Apr 2018 - Jul 2020
Github of the FaceForensics dataset
Contributions:64 commits, 44 pushes, 5 branches in 2 years 3 months
deepfakeface-alignmentdatasetface-detectionface
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.