Senior Physical AI Engineer
- AI
- Machine Learning
- Python
- C++
- AI/ML
- Computer Vision
About the Team & Mission
Senior Physical AI Engineer๋ ์ฐจ์ธ๋ ์์จ์ฃผํ์ ์ํ End-to-End(E2E) Planning Model๊ณผ ๋ชจ๋ธ์ ํ์ตยท๊ฒ์ฆ์ ์ํ Closed-loop Simulation ํ๊ฒฝ ๋ฐ ํ์ดํ๋ผ์ธ์ ์ค๊ณยท๊ฐ๋ฐ์ ์ฃผ๋ํฉ๋๋ค.
Trajectory Generation๊ณผ Decision-making์ ์ํํ๋ E2E Planning Model๋ถํฐ Closed-loop Simulation๊น์ง, ์ค์ ์ฐจ๋์ ์ฃผํ ์ฑ๋ฅ์ผ๋ก ์ด์ด์ง๋ ํต์ฌ ๊ธฐ์ ์ ๊ฐ๋ฐํฉ๋๋ค. Generative AI, Imitation Learning, Reinforcement Learning, 3D ํ๊ฒฝ ์ฌ๊ตฌ์ฑ ๋ฐ ๋ฌผ๋ฆฌ ๊ธฐ๋ฐ Simulation ๋ฑ ๋ค์ํ ๊ธฐ์ ์ ํ์ฉํ์ฌ ๋ชจ๋ธ๊ณผ ๊ฐ์ ํ๊ฒฝ์ ํจ๊ป ๋ฐ์ ์ํค๊ณ Sim-to-Real Gap์ ์ค์ ๋๋ค.
E2E Planning Model์ ์ค๊ณยท๊ฐ๋ฐ๋ถํฐ Closed-loop Simulation ๊ฒ์ฆ๊ณผ ์ค์ฐจ ์คํ, ๊ฒฐ๊ณผ ๊ธฐ๋ฐ ๋ชจ๋ธ ๊ฐ์ ๊น์ง ์ ์ฒด ๊ฐ๋ฐ ์ฌ์ดํด์ ์ฃผ๋ํฉ๋๋ค. Simulation๊ณผ ์ค์ฐจ ๊ฐ ์ฑ๋ฅ ์ฐจ์ด๋ฅผ ์ค์ด๊ณ , ๊ธฐ์ ์ ์์ฌ๊ฒฐ์ ๊ณผ ๋ฆฌ๋ทฐ๋ฅผ ํตํด ์ฐจ์ธ๋ Autonomous Driving System ๊ฐ๋ฐ์ ์ด๋๋๋ค.
The Senior Physical AI Engineer leads the design and development of end-to-end (E2E) planning models for next-generation autonomous driving, as well as closed-loop simulation environments and pipelines for training and validating them.
You will work across E2E planning models for trajectory generation and decision-making, closed-loop simulation, generative AI, imitation learning, reinforcement learning, 3D environment reconstruction, and physics-based simulation. You will help bridge the gap between virtual environments and real-world driving performance.
You will lead the full development cycleโfrom designing and developing E2E planning models to validating them in closed-loop simulation and real-vehicle experiments and improving them based on the results. You will reduce performance gaps between simulation and real vehicles and drive next-generation autonomous driving systems through technical decision-making and reviews.
Responsibilities
Trajectory Generation ๋ฐ Decision-making์ ์ํ E2E Planning Model ์ค๊ณยท๊ฐ๋ฐ ๋ฐ ์ฑ๋ฅ ๊ฐ์ ์ฃผ๋
E2E Planning Model์ ํ์ตยท๊ฒ์ฆ์ ์ํ Closed-loop Simulation ํ๊ฒฝ, ์๋๋ฆฌ์ค ๋ฐ ํ๊ฐ ์ฒด๊ณ ์ค๊ณยท๊ฐ๋ฐ ์ฃผ๋
E2E Planning Model๊ณผ Simulation ๋ฐ ์ค์ฐจ ์์คํ ๊ฐ ์ธํฐํ์ด์ค ํตํฉ๊ณผ ์ค์ฐจ ์คํ ์ฃผ๋
Simulation ๊ฒฐ๊ณผ์ ์ค์ฐจ ์ฃผํ ๋ฐ์ดํฐ ๋ฐ ์คํจ ์ฌ๋ก๋ฅผ ๋ถ์ํ๊ณ Sim-to-Real Gap์ ์ค์ด๊ธฐ ์ํ ๋ชจ๋ธ ๊ฐ์ ๋ฐฉํฅ ๋์ถ
๋ด๋น ์์ญ์ ๊ธฐ์ ์ ๋ฌธ์ ์ ๋ก๋๋งต์ ์ ์ํ๊ณ , ํ์ฅ์ฑยท์ฌํ์ฑยท์ด์ ์์ ์ฑ์ ๊ณ ๋ คํ ํต์ฌ ์์คํ ์ ์ํคํ ์ฒ ๋ฐ ๊ฐ๋ฐ ๋ฐฉํฅ ์ฃผ๋
์ค๊ณ ๋ฐ ์ฝ๋ ๋ฆฌ๋ทฐ, ๊ธฐ์ ๊ณต์ ์ ๋ฉํ ๋ง์ ํตํด ํ์ ์์ง๋์ด๋ง ํ์ง ํฅ์
Lead the design, development, and performance improvement of E2E planning models for trajectory generation and decision-making
Lead the design and development of closed-loop simulation environments, scenarios, and evaluation frameworks for training and validating E2E planning models
Lead interface integration between E2E planning models, simulation, and real-vehicle systems, as well as real-vehicle experiments
Analyze simulation results, real-world driving data, and failure cases to define model improvements that reduce the sim-to-real gap
Define technical problems and roadmaps, and lead the architecture and development of key systems with scalability, reproducibility, and operational reliability in mind
Raise the team's engineering quality through design and code reviews, knowledge sharing, and mentoring
Qualifications
์ปดํจํฐ๊ณตํ, ์ ์๊ณตํ, ์๋์ฐจ๊ณตํ, ๋ก๋ด๊ณตํ ๋๋ ๊ด๋ จ ๋ถ์ผ์ ํ์ฌ ํ์์ 8๋ ์ด์์ ์ค๋ฌด ๊ฒฝํ, ์์ฌ ์ด์ ํ์์ 5๋ ์ด์์ ์ค๋ฌด ๊ฒฝํ ๋๋ ์ด์ ์คํ๋ ์ญ๋
Python ๋๋ Modern C++ ๊ธฐ๋ฐ์ ๋ชจ๋ธ ๋ฐ Simulation ์์คํ ์ ์ค๊ณํ๊ณ ๊ฐ๋ฐยท๋๋ฒ๊น ํ ์ ์๋ ์ญ๋
Autonomous Driving, Robotics, AI/ML ๋๋ Simulation ๊ด๋ จ ์์คํ ์ ์ค๊ณ ๋ฐ ๊ฐ๋ฐ์ ์ฃผ๋ํ ๊ฒฝํ
๋ค์ ๋ถ์ผ ์ค ํ๋ ์ด์์ ๋ํ ๊น์ด ์๋ ์ ๋ฌธ์ฑ๊ณผ ์ด๋ฅผ ์ค์ ์์คํ ์ ์ ์ฉํ ๊ฒฝํ
๋จธ์ ๋ฌ๋ ๊ธฐ๋ฐ ์์จ์ฃผํ Planning Model ์ค๊ณ, ํ์ต ๋ฐ ์ฑ๋ฅ ๊ฐ์
Motion Planning, Decision-making ๋๋ Robotics ์๊ณ ๋ฆฌ์ฆ ์ค๊ณ, ์ฑ๋ฅ ๊ฐ์ ๋ฐ ์ค์ ์์คํ ์ ์ฉ
Computer Vision ๊ธฐ๋ฐ 3D ํ๊ฒฝ ์ฌ๊ตฌ์ฑ ๋ฐ ํํ
์ฐจ๋ยท๋ก๋ด์ ๋์ญํ ๋ชจ๋ธ๋ง ๋๋ Simulation ํ๊ฒฝ ์ค๊ณ ๋ฐ ๊ตฌ์ถ
์ฌ๋ฌ ์ง๊ตฐ์ ์ดํด๊ด๊ณ์์ ๊ธฐ์ ์ ์์ฌ๊ฒฐ์ ๋ฐ ์ฐ์ ์์๋ฅผ ์กฐ์จํ ์ ์๋ ์ปค๋ฎค๋์ผ์ด์ ์ญ๋
Bachelorโs degree in Computer Science, Electrical Engineering, Automotive Engineering, Robotics, or a related field with 8+ years of professional experience; or a masterโs degree or higher in one of these fields with 5+ years of professional experience; or equivalent practical expertise
Ability to design, develop, and debug model and simulation systems using Python or Modern C++
Experience leading the design and development of systems related to autonomous driving, robotics, AI/ML, or simulation
Demonstrated expertise inat least one of the following areas, with experience applying it to real-world systems:
Design, training, and performance improvement of machine learning-based planning models for autonomous driving
Design, performance improvement, and real-world system integration of motion planning, decision-making, or robotics algorithms
Computer vision-based 3D environment reconstruction and representation
Vehicle or robot dynamics modeling, or simulation environment design and development
Ability to communicate technical decisions and align priorities across cross-functional stakeholders
Preferred Qualifications
Autonomous Driving ๋๋ Robotics ๋ถ์ผ์์ E2E Planning Model๊ณผ Closed-loop Simulation์ ์ฐ๊ณํ ํ์ตยทํ๊ฐ Pipeline์ ์ค์ ์์คํ ์์ค์ผ๋ก ๊ฐ๋ฐยท์ด์ํ ๊ฒฝํ
์ค์ ์ฐจ๋ ๋๋ Robot์ E2E Planning Model์ ํตํฉยท๊ฒ์ฆํ๊ณ ์คํ์ ์ฃผ๋ํ ๊ฒฝํ
๋๊ท๋ชจ Simulation Pipeline ๋๋ Platform์ ์ค๊ณ, ์ฑ๋ฅ ์ต์ ํ ๋ฐ ์ด์ ๊ฒฝํ
๋ฐ์ดํฐยท๋ชจ๋ธยท์ฝ๋์ ๋ฒ์ ๊ด๋ฆฌ์ ์๋ํ๋ ํ ์คํธ ๋ฐ ํ๊ฐ๋ฅผ ํฌํจํ ์ฌํ ๊ฐ๋ฅํ ML ๊ฐ๋ฐ ํ๊ฒฝ์ ๊ตฌ์ถํ ๊ฒฝํ
ICRA, IROS, CVPR, NeurIPS, RSS ๋ฑ ๊ด๋ จ ํํ ๋๋ ์ ๋ ์ฐ๊ตฌยท๋ ผ๋ฌธ ๋ฐํ ๊ฒฝํ
Experience developing and operating production-scale training and evaluation pipelines integrating E2E planning models with closed-loop simulation for autonomous driving or robotics
Experience integrating and validating E2E planning models on real vehicles or robots and leading experiments
Experience designing, optimizing, and operating large-scale simulation pipelines or platforms
Experience building reproducible ML development environments with versioned data, models, and code, as well as automated testing and evaluation
Research or publication experience in relevant conferences or journals such as ICRA, IROS, CVPR, NeurIPS, or RSS
Interview Process
์๋ฅ ์ ํ
์ฝ๋ฉ ํ ์คํธ
1์ฐจ ๋ฉด์ (ํ์, 1์๊ฐ ๋ด์ธ)
2์ฐจ ๋ฉด์ (๋๋ฉด ํน์ ํ์, 3์๊ฐ ๋ด์ธ)
์ฒ์ฐ ํ์ยท์ ์ฌ
Application Screening
Coding Test
First Interview (Virtual, approximately 1 hour)
Second Interview (In-person or Virtual, approximately 3 hours)
Offer Discussion / Onboarding
Additional Information
์ ํ ์ ์ฐจ๋ ์ผ์ ๋ฐ ์งํ ์ํฉ์ ๋ฐ๋ผ ์ผ๋ถ ๋ณ๊ฒฝ๋ ์ ์์ผ๋ฉฐ, ๊ฐ ์ ํ ๊ฒฐ๊ณผ๋ ๋ฑ๋กํ์ ์ด๋ฉ์ผ๋ก ๊ฐ๋ณ ์๋ด๋๋ฆฝ๋๋ค.
์ง์์ ์ ์ถ ์ ์ฃผ๋ฏผ๋ฑ๋ก๋ฒํธ, ๊ฐ์กฑ๊ด๊ณ, ํผ์ธ ์ฌ๋ถ, ์ฐ๋ด, ์ฌ์ง, ์ ์ฒด์กฐ๊ฑด, ์ถ์ ์ง์ญ ๋ฑ ์ฑ์ฉ์ ์ฐจ๋ฒ์ ์๊ตฌ ๊ธ์ง๋ ์ ๋ณด๋ ์ ์ธ ๋ถํ๋๋ฆฝ๋๋ค.
์ง์์ ์ ์ ์ค ์ค๋ฅ๊ฐ ๋ฐ์ํ๊ฑฐ๋ ๊ธฐํ ๋ฌธ์ ์ฌํญ์ด ์์ ๊ฒฝ์ฐ, recruit@42dot.ai๋ก ๋ฌธ์ํด ์ฃผ์๊ธฐ ๋ฐ๋๋๋ค.
๊ตญ๊ฐ๋ณดํ๋์์ ๋ฐ ์ทจ์ ๋ณดํธ ๋์์๋ ๊ด๊ณ๋ฒ๋ น์ ๋ฐ๋ผ ์ฐ๋ํฉ๋๋ค.
์ฅ์ ์ธ ๊ณ ์ฉ ์ด์ง ๋ฐ ์ง์ ์ฌํ๋ฒ์ ๋ฐ๋ผ ์ฅ์ ์ธ ๋ฑ๋ก์ฆ ์์ง์๋ฅผ ์ฐ๋ํฉ๋๋ค.
42dot์ ์๋ขฐํ์ง ์์ ์์นํ์ ์ด๋ ฅ์๋ฅผ ๋ฐ์ง ์์ผ๋ฉฐ, ์์ฒญํ์ง ์์ ์ด๋ ฅ์์ ๋ํด ์์๋ฃ๋ฅผ ์ง๋ถํ์ง ์์ต๋๋ค.
์ง์์ ๋ด์ฉ ์ค ํ์ ์ฌ์ค์ด ๋ฐ๊ฒฌ๋ ๊ฒฝ์ฐ, ์ ์ฌ๊ฐ ์ทจ์๋ ์ ์์ต๋๋ค.
์ธํฐ๋ทฐ ํ๋ก์ธ์ค ์ข ๋ฃ ํ ์ง์์์ ๋์ํ์ ํํ์กฐํ๊ฐ ์งํ๋ ์ ์์ต๋๋ค.
3๊ฐ์์ ์์ต๊ธฐ๊ฐ์ด ์ ์ฉ๋ ์ ์์ต๋๋ค.
The recruitment process may change depending on schedule and progress; the result of each stage will be sent individually to your registered email.
Please do not include legally prohibited information in your application (e.g., ID number, family relations, marital status, salary, photo, physical details, hometown).
For application errors or inquiries, contact recruit@42dot.ai.
Veterans and applicants eligible for employment protection will receive preferential consideration in accordance with applicable laws and regulations.
In compliance with the Act on Employment Promotion and Vocational Rehabilitation for Persons with Disabilities, registered individuals with disabilities will receive preferential consideration.
42dot does not accept unsolicited resumes from search firms. We will not pay any fees for resumes submitted without prior agreement.
False information in your application may result in offer cancellation.
A reference check may be conducted after the interview process, with your consent.
A 3-month probationary period may apply.
Senior Physical AI Engineer ยท 42dot