Назад
Company hidden
8 дней назад

Physical Recursive Self Improvement (P-RSI)

Формат работы
onsite
Тип работы
fulltime
Грейд
trainee
Английский
b2
Страна
Switzerland
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/
TL;DR
Physical Recursive Self Improvement (P-RSI) (AI/Robotics): Developing a training system that improves reinforcement-learning pipelines through frequent sim-to-sim and periodic sim-to-real transfers, with an accent on recursive self-improvement, physical robotics, and learning from sparse real-robot trials. Focus on building general training rules, validating them across diverse simulated scenarios, and reducing the risks and costs of real-world experimentation.

Location: Zurich, Switzerland; onsite

Company

Research thesis conducted at the hirify.global in collaboration with ETH Zurich, using a mature robotics stack, multiple robots, engineering support, advanced computing infrastructure, and a robotics research laboratory.

What you will do

  • Explore recursive self-improvement for physical robotics and reinforcement-learning training pipelines.
  • Develop an AI system that improves task policies, algorithms, and environments through frequent sim-to-sim and periodic sim-to-real transfers.
  • Use sparse real-robot trials to derive general training rules rather than tuning the pipeline for a single task.
  • Validate and refine the rules across diverse simulated scenarios while minimizing physical risk.
  • Address challenges including partial observability, shifting dynamics, and the high cost of real-world failures.

Requirements

  • Master's thesis placement in the field of AI, robotics, or a related research area.
  • Work must be conducted onsite in Zurich as part of the hirify.global and ETH Zurich collaboration.
  • Ability to append a CV, letter of motivation, and transcripts to the application.

Culture & Benefits

  • Access to a mature robotics stack running on multiple robots and enabling fast testing cycles.
  • Strong engineering support and a state-of-the-art robotics research laboratory.
  • Access to advanced computing infrastructure and collaboration with distinguished robotics researchers.
  • Selection is highly competitive; shortlisted candidates are invited to an onsite interview.

Hiring process

  • Submit a CV, letter of motivation, and transcripts with the application.
  • Shortlisted candidates will be invited to an onsite interview.

Будьте осторожны

Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →