VLM research engineer

职责描述:

1. 在自动驾驶领域参与开发深度学习相关项目, 开发并优化面向自动驾驶的视觉语言模型（如多模态感知、场景理解等）

2. 针对实际场景需求（延迟、精度、效率等），实现并落地AI感知模型

3. 调研前沿视觉语言模型/自动驾驶技术（如E2E、VLA），完成技术选型评估。

4. 高质量地按计划完成项目开发任务。

能力需求:

1. 计算机科学、电子工程、机器人等相关专业硕士/博士（博士优先）

2. 专业能力：

• 必需项:

o 视觉语言模型/多模态模型实战经验（如CLIP、BEVFormer等）。

o 熟练使用Python及深度学习框架（PyTorch/TensorFlow）。

• 加分项:

o CVPR/ICCV等顶会论文发表经历。

3.熟练阅读/撰写英文技术文档，具备主动解决问题能力

4. 每周5天全职，实习期≥6个月，可尽快入职者优先

Responsibilities:

1. Develop and optimize vision-language models (e.g., multi-modal

perception, scene understanding) for autonomous driving systems.

2. Implement and deploy AI perception models under real-world constraints (latency, accuracy)

3. Research cutting-edge VLM/autonomous driving technologies (e.g., BEV, LLM integration) and evaluate feasibility

4. Hands-on development for the project with quality.

Requirements:

1. Master/PhD in CS, EE, Robotics, or related fields (PhD candidates preferred for VLM research).

2. Technical Skills

• Must-have:

Hands-on experience with VLM models (e.g., CLIP, BLIP, LLaVA).

Proficiency in Python and DL frameworks (PyTorch/TensorFlow).

• Preferred:

Publications at CVPR/ICCV/ECCV/IROS.

3. Language & Soft Skills

• Fluent in technical English (paper reading/writing).

• Strong problem-solving skills and self-driven mentality.

4. ≥6 months full-time (5 days/week), immediate onboarding preferred.