Research Intern: LLM Algorithms & Multimodal AI

ByteDance; San Jose (CA); On-site; USD 97,000 - 138,000; Part time

Benefits offered by this job

Health insurance

Life insurance

Wellbeing benefits

10 paid holidays

Paid sick time

Housing allowance

Job summary

ByteDance Seed invites PhD candidates to join its research team for a 12-week internship focused on foundational algorithm research for large models. You will work on LLM/MultiModal ML topics, contribute to papers and potential patents, and explore practical industry applications while collaborating with experienced researchers in the Seed group.

The program requires authorization to work in the United States and does not include immigration sponsorship.

Qualifications

• Pursuing a PhD in AI, CS, automation, math or related technical discipline.
• Strong foundation in data structures and algorithms.
• Proficient with Python and C++, and deep learning frameworks (PyTorch, TensorFlow).
• Experience with distributed large language model training frameworks.

Responsibilities

• Participate in R&D of cutting-edge algorithms with potential for international papers and patents.
• Conduct in-depth research on LLMs and MultiModal ML and apply findings to practical industry problems.

Skills

Python

C++

Data structures

Deep learning

Education

PhD candidate

Tools

Megatron

FSDP

Deepspeed

PyTorch

TensorFlow

Job description

ByteDance Seed invites PhD candidates to join its research team for a 12-week internship focused on foundational algorithm research for large models. You will work on LLM/MultiModal ML topics, contribute to papers and potential patents, and explore practical industry applications while collaborating with experienced researchers in the Seed group.

The program requires authorization to work in the United States and does not include immigration sponsorship.