Internship in Multimodal Embodied AI & 3D Scene Design

merl; Cambridge (MA); On-site; USD 28,000 - 41,000; Part time

Job summary

MERL is seeking a self-motivated intern to tackle problems at the intersection of multimodal large language models and embodied AI in dynamic indoor environments. The ideal candidate is a PhD student with a strong ML/vision background and top-tier publications.

Experience designing synthetic scenes using graphics software, and working with simulators like Habitat/SoundSpaces, SMPL models, and reinforcement learning is preferred.

Qualifications

• PhD-level candidate in ML/vision required.
• Experience designing 3D interactive scenes.
• Strong publication record in CVPR/NeurIPS or similar venues.
• Experience with simulators Habitat/SoundSpaces and SMPL models.

Responsibilities

• Collaborate with MERL computer vision researchers to develop algorithms.
• Prepare manuscripts for scientific publications.
• Conduct experiments in dynamic indoor environments and report results.

Skills

3D scene design

Vision-based embodied AI

Multimodal LLMs

Reinforcement learning

Programming fundamentals

Publications in top venues

Education

PhD student

Tools

Habitat

SoundSpaces

SMPL

Job description

MERL is seeking a self-motivated intern to tackle problems at the intersection of multimodal large language models and embodied AI in dynamic indoor environments. The ideal candidate is a PhD student with a strong ML/vision background and top-tier publications.

Experience designing synthetic scenes using graphics software, and working with simulators like Habitat/SoundSpaces, SMPL models, and reinforcement learning is preferred.