Marcos Pividori
Software Engineer at Google
Zurich, Zurich, Switzerland
Summary
🤩
Rockstar🎓
Top SchoolMarcos Pividori is a software engineer with 13 years of experience, currently focused on video abuse detection for YouTube Trust & Safety at Google in Zurich. He brings deep systems and C/C++ expertise from long-standing contributions to high-performance open-source projects like mlpack (fast C++ ML library) and OpenCog, where he fixed core algorithmic bugs and built Haskell bindings that carefully manage memory and exceptions. His background includes fuzzing and LLVM work, robotics firmware at iRobot, and multiple Google internships, reflecting a strong foundation in low-level engineering and applied machine learning. As a former university teaching assistant, he pairs engineering rigor with clear technical communication and a knack for refactoring complex codebases into safer, more maintainable abstractions. An interesting detail: he’s implemented spill trees for approximate nearest-neighbor search and authored monad-based Haskell APIs that bridge to C++ graph databases.
12 years of coding experience
1 year of employment as a software developer
Licenciatura en Ciencias de la Computación, Computer Science, Licenciatura en Ciencias de la Computación, Computer Science at Universidad Nacional de Rosario - UNRInstituto Politécnico Superior Gral. San Martín
English, Spanish
Github Skills (16)
algorithm10
data-structures10
algorithms10
machine-learning10
c-language10
monads10
cprogramming-language10
haskell10
data-structure10
ffi10
ff10
template-haskell9
nearest-neighbors9
nearest-neighbor-search9
knowledge-representation8
Programming languages (6)
Github contributions (5)
mlpack/mlpack
Mar 2016 - Aug 2016
mlpack: a fast, header-only C++ machine learning library
Role in this project:
Back-end Developer
Back-end DeveloperContributions:191 commits, 27 PRs, 1 push in 5 months
Contributions summary:Marcos contributed to the core C++ machine learning library, mlpack, by addressing errors in existing code. Their commits focused on fixing incorrect calculations within the neighbor search algorithms and correcting mistakes in metric policy. The user also properly implemented and refactored use of the Enum type throughout the library.
regressionheaderdeep-learningscientific-computingc-plus-plus
opencog/atomspace
Mar 2015 - Sep 2015
The OpenCog (hyper-)graph database and graph rewriting system
Role in this project:Back-end Developer
Back-end DeveloperContributions:143 commits, 21 PRs, 48 comments in 5 months
Contributions summary:Marcos primarily focused on developing Haskell bindings for the OpenCog AtomSpace project. Their contributions involved implementing a monad-based API for interacting with the C++ AtomSpace class, enabling the creation and management of atoms. They also refactored the code to use the ReaderT monad transformer and addressed memory management issues, ensuring proper object disposal and exception handling. Further work included adding new atoms representation and pattern matching features.
logic-programmingopencogknowledge-baseedgedbknowledge-graph
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.