AI & Robotics Research Internship — LLMs, Vision & Control

Amazon Science; San Francisco (CA); On-site; USD 134,000 - 164,000; Full time

Job summary

Amazon Science in San Francisco, CA invites an intern to explore the frontier of intelligent robotics where ML, deep learning, and robotics converge. You will work alongside world‑renowned researchers, gaining hands‑on experience with state‑of‑the‑art computing infrastructure and cutting‑edge projects at the intersection of AI and robotics.

You will help develop scalable algorithms, tackle challenging research on perception, manipulation, and control, and collaborate with cross‑functional teams

Qualifications

• Enrolled in a Bachelor's degree or higher and eligible for full-time internship.
• Must be willing to relocate to internship base.
• Experience programming in Python; C++ or JAX optional.
• Demonstrated depth in a research area via publications, thesis, or equivalent.
• Available for the full duration of the internship.

Responsibilities

• Develop scalable algorithms and modeling techniques at the intersection of LLMs and robotics.
• Tackle production-scale research problems in perception, manipulation and control.
• Collaborate with cross-functional teams and clearly communicate findings.
• Work independently in a fast-paced, evolving environment with diverse stakeholders.

Skills

Machine learning

Deep learning

Robotics

Multimodal LLMs

World models

Image/video tokenization

Vision-language-action models

Python

PyTorch

JAX

Problem solving

Education

Bachelor's degree or higher

Relocation to internship location

Full-time internship availability

Publications/thesis evidence

Tools

MuJoCo

Isaac Sim

Job description

Amazon Science in San Francisco, CA invites an intern to explore the frontier of intelligent robotics where ML, deep learning, and robotics converge. You will work alongside world‑renowned researchers, gaining hands‑on experience with state‑of‑the‑art computing infrastructure and cutting‑edge projects at the intersection of AI and robotics.

You will help develop scalable algorithms, tackle challenging research on perception, manipulation, and control, and collaborate with cross‑functional teams