We are seeking an experienced Quantitative Developer to join the Numerical Performance Group (NPG), a central specialist team within Citis Markets Quantitative Analysis (MQA) organisation.

NPG designs, develops, and deploys roots, Citis core highperformance C++ numerical library. The roots library underpins pricing and risk infrastructure used across multiple assetclass quantitative teams and is engineered for maximum accuracy and performance on modern hardware.

The team works closely with frontoffice quantitative groups and trading desks, tackling critical performance, scalability, and stability challenges across Citis derivatives pricing stack.

Responsibilities

• Design, develop, and enhance quantitative libraries used for pricing and risk management
• Create, implement, and support quantitative models for the trading business using advanced mathematical and computational techniques
• Apply highperformance computing methods, including hardware acceleration and lowlevel optimisation
• Develop pricing models using numerical techniques such as Monte Carlo methods and partial differential equation (PDE) solvers
• Work with technologies including C++, CUDA, Python, and adjoint algorithmic differentiation (AAD)
• Contribute to the technical direction of the group, mentor junior team members, and collaborate closely with quant teams across asset classes

Skills and Experience

• Proven experience in a highperformance computing or numerical software role
  (experience outside of finance will be considered)
• Strong programming skills in C++; experience with CUDA and Python preferred
• Excellent background in computational mathematics, numerical analysis, or a related quantitative discipline
• Demonstrated ability to design, implement, and optimise complex mathematical algorithms for performancecritical applications
• Solid understanding of Adjoint Algorithmic Differentiation (AAD) concepts; handson experience with AAD tools is highly desirable
• Deep practical knowledge of lowlevel optimisation techniques, including SIMD intrinsics, autovectorisation, cache behaviour, and memory access patterns
• Strong understanding of modern hardware architectures and their impact on computational performance
• Experience developing and optimising software on both Windows and Linux
• Clear, concise written and verbal communication skills, with the ability to collaborate effectively across teams

Education

Candidates should hold a postgraduate degree in a numerate discipline such as Mathematics, Physics, Computer Science, Engineering, or a related field.

Given the seniority and specialist nature of the role, a PhD is strongly preferred.