This job might no longer be available.

Numerical Optimization Researcher

Niantic’s Mapping (R&D) Team is seeking an ambitious Numerical Optimization Researcher to push the frontiers of Augmented Reality, and scalable optimization in particular.

Niantic Engineering leads the advancement of AR and other immersive technologies, while crafting engaging apps for a user base in the billions. You will work as part of a fun team, to invent and help develop novel algorithms and systems that sit at the cross-section of optimization, computer vision, machine learning, and robotics.

We truly believe that the company performs at its best when our employees are encouraged to work on what they enjoy. Our team culture means we’re focussed on sharing knowledge between all of our engineers which is why we hold a lot of great technical talks on a weekly basis.

We are passionate about discovery, so we’re looking for scientists who want to learn, publish, and deploy every new invention that could help make better “adventures on foot, with others.”

Responsibilities

• Develop flexible non-linear least squares solvers for addressing the company’s vision problems.
• Design and deploy solvers to perform bundle adjustment in the context of large-scale structure from motion.
• Collaborate with team members in Mapping, R&D, and beyond, to build, test, and improve software systems based on empirical evidence.
• Keep abreast of emerging best-practices in mathematical programming and optimization theory, and communicate these to others.

Qualifications

• You have a PhD or PhD-level qualifications in either Mathematics, Physics, Electrical Engineering, or Computer Science.
• You have demonstrable experience developing and deploying custom solvers for large- scale constrained optimization problems.
• You have hands on programming experience with python, numpy and scipy.
• You have training in mathematical/numerical optimization, matrix calculus, advanced linear algebra, functional analysis, numerical analysis and approximation theory.
• You have experience with designing and developing iterative solvers for parallelizable computational resources.

Plus If...

• You have experience developing and deploying custom solvers for optimization of non- linear (robust) least squares problems with sparse Jacobians.
• You have experience developing inexact Levenberg-Marquardt solvers for large sparse non-linear least squares problems (see Wright and Holt, Australian Mathematical Society 1985).
• You have programming experience with C++, CUDA, cuBLAS, Ceres.
• You have programming experience with cupy.
• You’re familiar/have experience with SLAM systems or large-scale bundle adjustment problems in structure-from-motion setting.
• You have recent publications on mathematical optimization in mathematical optimization venues such as Journal Optimization, Journal of Optimization Theory and Applications, Optimization Letters, Optimization and Engineering, Optimization Methods and Software, Mathematical Programming, Mathematical Programming Computation, Journal on Scientific Computing, etc.
• You have publications on structure-from-motion, bundle adjustment or SLAM in computer vision venues such as CVPR, ECCV, ICCV, IJCV, WACV, PAMI, etc.
• You are familiar with rotation-averaging approaches for large-scale structure-from-motion problems.
• You have coded contributions to Google’s Ceres Solver or similar open-source solvers.

Join the Niantic team!

Niantic is the world’s leading AR technology company, sparking creative and engaging journeys in the real world. Our products inspire outdoor exploration, exercise, and meaningful social interaction.

Originally formed at Google in 2011, we became an independent company in 2015 with a strong group of investors including Nintendo, The Pokémon Company, and Alsop Louie Partners. Our current titles include pioneering global-control game Ingress, record-breaking AR game Pokémon GO, and recently released third title, Harry Potter: Wizards Unite.

Niantic is an Equal Opportunity and Affirmative Action employer. We believe that cultivating a workplace where our people are supported and included is essential to creating great products our community will love. Our mission emphasizes seeking and hiring diverse voices, including those who are traditionally underrepresented in the technology industry, and we consider this to be one of the most important values we hold close.

We're a hard-working, fun, and exciting group who value intellectual curiosity and a passion for problem-solving! We have growing offices located in San Francisco, Sunnyvale, Bellevue, Los Angeles, London, Tokyo, Hamburg, and Zurich.