Senior AI Research Scientist - LLMs & Agent Tech

Perplexity; San Francisco, Palo Alto (CA, CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Perplexity is seeking top-tier AI Research Scientists and Engineers to advance our AI products and capabilities. We are building the future of AI-powered search and agent experiences through our Sonar models, Deep Research Agent, Comet Agent, and Search products.

Join us in creating SOTA experiences that handle hundreds of millions of queries and continue to scale rapidly. We offer collaborative teams across Core Research, Agent Products, and Comet Agent, focusing on model training,

Qualifications

• Proven experience with large-scale LLMs and Deep Learning systems.
• Strong programming in Python and PyTorch; versatility a plus.
• Experience with post-training techniques and reinforcement learning.
• Self-starter with ownership of tasks.
• 2–6 years of relevant experience depending on seniority.

Responsibilities

• Post-train SOTA LLMs using supervised and reinforcement learning techniques.
• Scale model performance across Sonar, Deep Research, Comet, and Search products.
• Stay current with LLM research, training, optimization, and personalization.
• Implement preference optimization and personalization to improve UX.
• Turn research ideas into algorithms and run experiments to launch new models.
• Own full-stack data, training, and evaluation pipelines for model development.

Skills

Python

PyTorch

LLMs

Reinforcement Learning

Post-training

Tools

Megatron

CUDA

Training frameworks

Job description

Perplexity is seeking top-tier AI Research Scientists and Engineers to advance our AI products and capabilities. We are building the future of AI-powered search and agent experiences through our Sonar models, Deep Research Agent, Comet Agent, and Search products.

Join us in creating SOTA experiences that handle hundreds of millions of queries and continue to scale rapidly. We offer collaborative teams across Core Research, Agent Products, and Comet Agent, focusing on model training,