![Banseog [ physical ai ]](/assets/brand/banseog-physical-ai-on-dark.svg?v=20261007){kind=small}
At a glance
Engineer for learning-based autonomous navigation algorithm R&D
What you will do
- Develop and improve RL and IL based navigation policies
- Train and evaluate models using simulation and real robot data
- Integrate learning-based and traditional autonomous driving systems
What they ask for
- Degree or equivalent experience in CS, robotics, AI
- Deep understanding and implementation experience of RL or IL
- Experience with Python and PyTorch for model development
Nice to have
- Master's/PhD or 3+ years R&D experience
- Experience in RL/IL navigation policy development for mobile robots
- Experience in Embodied AI research and development
Summarised by Banseog from the company's official posting. The official posting is the reference for conditions and details.
What the posting states
Only what the company's official posting says, in its own words.
- Education
- Master's · PhDPreferred
Not stated in the posting: Experience · Work mode · Pay · Visa sponsorship · Language. Not stated does not mean not offered. Check the official posting.
Skills in the posting
Planning roles: the market now
From company career sites · updated daily
Hiring most
Most requested skills
Countries with most jobs
- South Korea
- 12
- United States
- 2
- India
- 2
- France
- 1
Observed numbers only. No interpretation or recommendation.
Similar jobs
See all jobs →Same role · Planning
자율주행 연구원Raion Robotics · South Korea[Humanoid Robot] Perception & SLAM 엔지니어WIRobotics · 서울 사무소 (Robot Innovation Hub) · South KoreaExperience 3+ yearsVSLAM Engineer - Team Lead (Visual Positioning Team)STRADVISION · 서울 · South KoreaSr. Path Planning Algorithm Engineer (Multi-Camera Vision Path Planning)STRADVISION · 서울 · South KoreaExperience 7+ years자율주행팀 자율주행SW 엔지니어Rainbow Robotics · 세종 · South Korea휴머노이드 네비게이션 엔지니어 모집ROBOTIS · 대한민국 서울특별시 강서구 마곡중앙5로1길 37 · South Korea
Banseog checks official company careers pages every day. A job is marked closed only when the official page returns a removal (404 or 410).