Reinforcement Learning Engineer (Aerologistics Center)

We are forming a development center for autonomous UAVs and onboard systems. The team works at the intersection of robotics, computer vision, control systems, embedded development, communications, and flight testing. Our task is to translate research hypotheses into engineering solutions that can be verified on a simulator, test bench, and real platform.

We need specialists who can work with uncertainty but deliver results with measurable quality: formulate metrics, write reliable code, analyze telemetry, document interfaces, and participate in testing. We value experience with autonomous systems, UAVs, robotics, ADAS, industrial vision, embedded/Linux, and safety-critical development.

The projects are characterized by a high degree of autonomy and offer the opportunity to influence technical and research decisions.

Responsibilities

• Participation in R&D projects with industrial and academic partners of the aerologistics center
• Conducting applied research in the field of RL, optimal control, and related methods
• Development, training, and experimental evaluation of models (RL / DL)

Requirements

• Deep practical knowledge of Python and experience with the ML/DL stack (PyTorch, NumPy, pandas, sklearn, wandb; experience with JAX is a plus)
• Deep understanding of ML/DL algorithms. Experience in training and implementing RL algorithms from scratch
• Experience in developing and implementing ML/DL/RL solutions in applied or industrial settings, understanding of model quality assessment principles
• Independence, responsibility, ability to formalize tasks and deliver solutions to completion

Conditions

• Office format in Moscow (Skoltech)
• Annual salary review and annual bonus
• Extended medical insurance and preferential insurance for family
• Sber's unique training system for professional and career development
• Favorable mortgage for employees
• Free SberPrime+ subscription, discounts on partner company products
• Referral bonus for recommending friends to the Sber team
• Corporate pension program.