Multimodal AI Research Intern: RL Tuning & Responsible AI

Sony AI; United States; Remote; USD 150,000 - 190,000; Full time

Job summary

Sony AI is Sony's research organization applying AI to unleash human creativity across gaming, entertainment, and music divisions. The Ethical Creative and Protective AI Lab operates as a globally distributed team across the U.S. and Canada, with about 900 million Sony devices in use worldwide.

We seek a researcher pursuing or holding a PhD in CS/ML with hands-on RL fine-tuning of LLMs/VLMs to contribute to post-training infrastructure and advanced multimodal AI evaluation.

Qualifications

• Pursuing or completed a PhD in CS/ML.
• Hands-on RL fine-tuning of LLMs/VLMs preferred.

Responsibilities

• Contribute to post-training and evaluation infrastructure for multimodal AI systems.
• Fine-tune LLMs/VLMs using SFT, RLAIF, and RLVR.

Skills

RL fine-tuning experience

Education

PhD in CS/ML

Job description

Sony AI is Sony's research organization applying AI to unleash human creativity across gaming, entertainment, and music divisions. The Ethical Creative and Protective AI Lab operates as a globally distributed team across the U.S. and Canada, with about 900 million Sony devices in use worldwide.

We seek a researcher pursuing or holding a PhD in CS/ML with hands-on RL fine-tuning of LLMs/VLMs to contribute to post-training infrastructure and advanced multimodal AI evaluation.