Algorithms Engineer, Differential Privacy

As an Algorithms Engineer at Oblivious, a start-up focused on privacy-enhancing technologies, you will play a crucial role in designing and implementing core components of differential privacy systems. Your primary responsibility will be to translate mathematical theory into production-ready code, with a focus on making privacy guarantees practical and efficient. Key Responsibilities: - Implement and analyze privacy loss accountants (RDP, zCDP) and their conversions to (, )-DP. - Develop algorithms for static and dynamic sensitivity analysis of relational operators for the Differentially Private SQL Engine. - Utilize Python AST manipulation and static analysis to enforce a DP-safe execution environment for Compiler & Static Analysis. - Implement and benchmark state-of-the-art algorithms for high-dimensional synthetic data generation for DP Synthetic Data. Qualifications Required: - Strong foundation in probability, statistics, and linear algebra. - Proficiency in Python for scientific computing, including numerical stability considerations. - Ability to translate mathematical concepts into robust, well-tested code. Desirable Skills: - Direct experience with differential privacy concepts or libraries. - Knowledge of compiler design, abstract syntax trees, or program analysis. - Experience with machine learning, particularly with noise models and generative models. - Familiarity with SQL parsers or database internals. Oblivious is looking for individuals who are passionate about privacy-enhancing technologies and want to be part of a team that is changing the landscape of data security and collaboration. If you are ready for a new challenge and an opportunity to advance your career in a supportive work environment, don't hesitate to apply today and join us in our mission to make privacy-preserving technologies the new norm. As an Algorithms Engineer at Oblivious, a start-up focused on privacy-enhancing technologies, you will play a crucial role in designing and implementing core components of differential privacy systems. Your primary responsibility will be to translate mathematical theory into production-ready code, with a focus on making privacy guarantees practical and efficient. Key Responsibilities: - Implement and analyze privacy loss accountants (RDP, zCDP) and their conversions to (, )-DP. - Develop algorithms for static and dynamic sensitivity analysis of relational operators for the Differentially Private SQL Engine. - Utilize Python AST manipulation and static analysis to enforce a DP-safe execution environment for Compiler & Static Analysis. - Implement and benchmark state-of-the-art algorithms for high-dimensional synthetic data generation for DP Synthetic Data. Qualifications Required: - Strong foundation in probability, statistics, and linear algebra. - Proficiency in Python for scientific computing, including numerical stability considerations. - Ability to translate mathematical concepts into robust, well-tested code. Desirable Skills: - Direct experience with differential privacy concepts or libraries. - Knowledge of compiler design, abstract syntax trees, or program analysis. - Experience with machine learning, particularly with noise models and generative models. - Familiarity with SQL parsers or database internals. Oblivious is looking for individuals who are passionate about privacy-enhancing technologies and want to be part of a team that is changing the landscape of data security and collaboration. If you are ready for a new challenge and an opportunity to advance your career in a supportive work environment, don't hesitate to apply today and join us in our mission to make privacy-preserving technologies the new norm.