Ondřej Cífka
Research Scientist at AudioShake
Southampton, England, United Kingdom
Summary
👤
Senior🎓
Top SchoolOndřej Cífka is a research scientist with 11 years of experience applying Transformer-based and generative models to speech, NLP and music processing, currently working at AudioShake in Southampton. He holds a PhD in AI and has a track record of academic and industry research—from Google and InterDigital internships to postdoctoral work on transformer explainability—bridging rigorous papers with pragmatic engineering. Ondřej contributes to prominent open-source projects like Magenta and Magenta.js, improving music generation pipelines and browser visualizers, reflecting both backend ML and full-stack skills. His work includes innovations in efficient positional encodings and one-shot music style transfer, and he uniquely combines formal computational linguistics training with jazz studies, informing creative approaches to audio and sequence modeling.
11 years of coding experience
2 years of employment as a software developer
Jazz, Jazz at VOŠ Jaroslava JežkaPhD, Computer Science, Data, AI, PhD, Computer Science, Data, AI at Institut Polytechnique de ParisBachelor's degree, Computer Science, Bachelor's degree, Computer Science at Charles University in Prague
Github Skills (14)
music-generation10
typescript10
machine-learning10
javascript10
typescripts10
web-development10
tensorflow10
python10
typescript-types10
deep-learning9
svg9
css8
html8
build-process7
Programming languages (13)
Github contributions (5)
magenta/magenta
Apr 2019 - Nov 2019
Magenta: Music and Art Generation with Machine Intelligence
Role in this project:
ML Engineer
ML EngineerContributions:18 commits, 6 PRs, 26 comments in 6 months
Contributions summary:Ondřej primarily focused on refactoring and moving code related to sequence example creation and event encoding/decoding within the Magenta music generation framework. They updated and moved functions such as `make_sequence_example` and `EncoderPipeline` to improve modularity. Furthermore, the user adjusted imports and updated test files, suggesting a focus on streamlining the code base and improving maintainability for machine learning pipelines.
artmusic-generationgenerative-artmagentamachine-learning
magenta/magenta-js
Feb 2020 - Jun 2021
Magenta.js: Music and Art Generation with Machine Learning in the browser
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:11 commits, 14 PRs, 36 comments in 1 year 4 months
Contributions summary:Ondřej focused on enhancing the user interface and core functionality of the Magenta.js library. They implemented new features in the SVG piano roll visualizer, incorporating data attributes and making code improvements. Furthermore, the user addressed bugs related to player playback and seeking, improved the visualizer with CSS integration, and refined the codebase by optimizing sequences and building processes.
artbrowsermagenta-jsjavascriptmagenta
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.