- どんな仕事か
-
Developing 2D/3D object detection models to identify and estimate the position, size, orientation, and velocity of road users and obstacles.
Building online map perception algorithms to detect lanes, road boundaries, crosswalks, and stop lines.
Developing 3D occupancy models to estimate occupied, free, and unknown space.
Creating 3DGS-based scene reconstruction and rendering pipelines using camera and LiDAR data.
Developing camera, LiDAR, and multimodal perception models, including BEV and sensor fusion.
Building data and training pipelines using large-scale real-world driving data.
Conducting model training, debugging, evaluation, and performance optimization across challenging driving scenarios.
Analyzing model errors and improving datasets and algorithms.
Collaborating with Prediction, Planning, Engineering, and Infrastructure teams to support system integration and deployment.
- 求められるスキルは
-
必須 Hands-on experience in 2D/3D object detection, online map perception, 3D occupancy prediction, or 3DGS-based scene reconstruction.
Strong Python or C++ programming skills and practical experience with PyTorch or similar deep learning frameworks.
Solid understanding of object detection, model training, evaluation metrics, and post-processing.
Knowledge of 3D geometry, camera projection, sensor calibration, coordinate transforms, and camera/LiDAR data.
Understanding of map perception, vector representations, occupancy models, and 3D scene reconstruction.
Ability to conduct experiments, analyze complex failure cases, and improve model performance.
Experience maintaining reproducible workflows and collaborating on engineering integration. - 雇用形態は
- Full Time / 正社員
- どこで働くか
- 東京都
- 勤務時間は
- フレックスタイム制(コアタイムあり)
- 給与はどのくらい貰えるか
- 1000万円~1999万円
- 待遇・福利厚生は
- Employment Insurance / 雇用保険,Industrial Accident Compensation Insurance / 労災保険,Employees' Pension Insurance / 厚生年金保険,Employees' Health Insurance / 健康保険
- 休日休暇は
- [勤務時間]
フレックスタイム制(コアタイム あり)
リモートワーク :要相談
[休日・休暇]
土・日・祝日
年次有給休暇
慶弔休暇、産休・育休等"
NEW
掲載期間26/11/05~26/11/23
求人No.TC-28251





