Senior Multimodal AI Scientist - Agentic RL & LLMs

NVIDIA Corporation; Santa Clara (CA); On-site; USD 152,000 - 288,000; Full time

Benefits offered by this job

Equity

Comprehensive benefits

Job summary

NVIDIA Corporation in California seeks a Senior Deep Learning Scientist to advance multimodal AI capabilities across Nemotron Omni and VoiceChat. You will contribute to developing and deploying large language models for agentic systems, enabling audio-visual reasoning, tool use, and document understanding.

You will apply foundational research, improve post-training and alignment methods, and collaborate to scale experiments, data pipelines, and evaluation benchmarks in a fast-paced environment.

Qualifications

• Master’s degree or PhD in CS/AI/Applied Math with 5+ years of relevant experience.
• Excellent Python programming skills with strong fundamentals in scalable model development.
• Strong knowledge of ML/DL techniques and foundation model architectures (Transformers, MoE).
• Foundational understanding of reinforcement learning algorithms and implementation.

Responsibilities

• Apply research to develop, train, fine-tune, and deploy large language models for agentic systems (audio-visual reasoning, tool usage, document understanding).
• Advance post-training and alignment methods (instruction tuning, preference optimization, RLHF/RLVR/MOPD).
• Research agentic reasoning and grounded perception for planning and long-horizon tasks in digital/physical environments.
• Lead data collection, development, and benchmarking of multimodal datasets with rigorous evaluation.

Skills

Python

PyTorch

Deep learning

Education

Master’s degree or PhD in Computer Science, AI, or Applied Math

Tools

Transformers

Job description

NVIDIA Corporation in California seeks a Senior Deep Learning Scientist to advance multimodal AI capabilities across Nemotron Omni and VoiceChat. You will contribute to developing and deploying large language models for agentic systems, enabling audio-visual reasoning, tool use, and document understanding.

You will apply foundational research, improve post-training and alignment methods, and collaborate to scale experiments, data pipelines, and evaluation benchmarks in a fast-paced environment.