Ever wanted to join a vibrant young start-up? To tangibly change the world for the better?
Oblivious builds privacy-enhancing technologies to help organisations unlock insights from sensitive data. We are recruiting an Algorithms Engineer to design and implement the core components of our differential privacy (DP) systems, including our Private Python runtime, DP-SQL engine, and synthetic data generator.
This role requires translating mathematical theory into production-ready code. You will work on the fundamental challenges of making rigorous privacy guarantees practical and efficient.
Who We Are: Oblivious is a start-up focused on enabling secure data collaboration through privacy-enhancing technologies. We were founded by two former PhDs in machine learning and cryptography from the University of Oxford who are on a mission to make privacy-preserving technologies the new norm across the industry. We are backed by some of the most well-respected VCs in Europe and the US, and we are putting together a core product and development team. You will get to build platforms that are leveraged by the largest financial institutions and telecoms companies in the world.
Responsibilities
- Design, implement and evaluate privacy-preserving algorithms.
- Implement and integrate various differential privacy mechanisms.
- Calibrate and apply noise mechanisms (Gaussian, Laplace) based on rigorous sensitivity analysis.
- Use Python AST manipulation and static analysis to enforce a DP-safe execution environment, ensuring user-submitted code cannot leak private information.
- Privacy Accounting & Mechanisms: Analyse and implement privacy accounting and mechanisms.
- DP Synthetic Data: Implement and benchmark state-of-the-art algorithms for high-dimensional synthetic data generation.
- Collaborate with cross-functional teams to design, develop, and deploy privacy-preserving systems.
Requirements
- Strong foundation in probability, statistics, and linear algebra. You must be comfortable with statistical modelling, proving bounds, and reasoning about error/variance.
- Proficiency in Python for scientific computing, including numerical stability considerations (e.g., floating-point precision, clipping, scaling).
- Demonstrated ability to translate mathematical concepts from academic papers or technical specifications into robust, well-tested code.
- Excellent knowledge of compiler engineering, ideally having built a simple compiler in the past.
- Experience with data structures and ASTs.
Desirable
- Experience in designing and implementing privacy-preserving algorithms
- Experience with machine learning, particularly with noise models, statistical learning theory, or generative models.
- Familiarity with SQL parsers or database internals.
Join our team of talented and motivated individuals who are passionate about making a difference in the world of privacy-enhancing technologies. Apply now and be part of an innovative and exciting start-up that is revolutionising the industry!
Benefits
- Private Health Insurance
- Paid Time Off
- Work From Home, with one required in-office anchor week every six weeks for deep collaboration and planning
- Training & Development
