Carles Cladellas
Founder at Precognit
Sant Llorenç Savall, Catalonia, Spain
Summary
🤩
Rockstar🎓
Top SchoolCarles Cladellas is a founder and machine learning software engineer from Catalonia with 12 years of experience building open-source tooling and production ML systems. A former researcher at MIT’s DAI-Lab, he led development on Synthetic Data Vault (SDV) components and Automated Machine Learning libraries that were used in DARPA D3M projects and published at SIGMOD. He co-founded DataCebo and Pythia, scaling engineering teams and turning lab research into widely adopted synthetic-data tooling, and now leads Precognit. Technically fluent across Python, CI/CD, backend engineering and model-driven data transforms, his contributions span core GAN and copula implementations, CI workflows, and secure distributed systems. Based in Sant Llorenç Savall, he combines academic rigor with hands-on full-stack delivery—often improving project structure, dependency hygiene, and reproducibility behind the scenes.
12 years of coding experience
14 years of employment as a software developer
Licentiate degree Mathematics, Licentiate degree Mathematics at Universitat Autònoma de BarcelonaShodan: 1D Certified Penetration Tester Seguridad informática y de sistemas, Shodan: 1D Certified Penetration Tester Seguridad informática y de sistemas at HackingDojo.com
Catalan, Spanish, English, Portuguese, German
Github Skills (42)
data-generation10
dependency-management10
pytorch10
docker10
esb10
python10
packaging10
machine-learning10
datatable10
codecov10
generative-adversarial-network10
dockers10
cicd10
packing10
data-transformation10
Programming languages (8)
Github contributions (5)
sdv-dev/Copulas
Jun 2018 - Jan 2022
A library to model multivariate data using copulas.
Role in this project:
Back-end Developer
Back-end DeveloperContributions:11 releases, 46 reviews, 208 commits in 3 years 7 months
Contributions summary:Carles primarily contributed to the library's underlying structure and dependency management. They reorganized project dependencies within the `setup.py` file, updating requirements and configuration. Additionally, the user updated documentation references to point to the current repository and moved examples to a separate folder. These changes indicate a focus on improving the project's maintainability, usability, and overall structure.
pythontabular-datagenerative-modelmachine-learningdataset-generation
sdv-dev/CTGAN
Nov 2019 - Jan 2022
Conditional GAN for generating synthetic tabular data.
Role in this project:Full-stack Developer
Full-stack DeveloperContributions:6 releases, 47 reviews, 89 commits in 2 years 3 months
Contributions summary:Carles upgraded the project's cookiecutter and fixed linting issues, indicating a focus on project setup and code quality. They then implemented and maintained crucial features within the `ctgan/model.py` and `ctgan/transformer.py` files. These changes involve the core of the CTGAN, including Gaussian Mixture Models, data transformation, and applying activation functions for data generation, showcasing a deep understanding of the project's core functionality and data generation. Further commits included code reorganization and ensuring support for categorical features, demonstrating a commitment to improving the software's structure and capabilities.
pytorchdeep-learningconditionaltabular-datagenerative-adversarial-network
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.