AI Researcher
Advances AI products through post-training SOTA LLMs using supervised and reinforcement learning techniques on rich query datasets. Owns data pipelines, training frameworks, and model integration while collaborating across teams. Requires 2-6+ years in large-scale LLMs and Python/PyTorch expertise; PhD preferred.
About the job
Responsibilities
Research & Development
- Post-train SOTA LLMs using the latest supervised and reinforcement learning techniques (SFT/DPO/GRPO)
- Leverage our rich query/answer dataset to scale model performance across Sonar, Deep Research, Comet, and Search products
- Stay current with the latest LLM research, especially in model training, optimization, and personalization techniques
- Implement preference optimization and personalization capabilities to enhance user experience
- Invent in-house improvements and optimizations to enhance SOTA models
- Turn research ideas into algorithms and run experiments to launch new models
Infrastructure & Implementation
- Own full-stack data, training, and evaluation pipelines required for model development
- Build robust and effective training frameworks (on top of Megatron/PyTorch) for post-training LLMs
- Implement necessary infrastructure and components to support cutting-edge model training at scale
- Integrate models seamlessly into our product ecosystem
Collaboration
- Work closely with engineering teams to integrate models into Perplexity's product suite
- Collaborate across teams to ensure cohesive AI experiences throughout our platform
- Partner with product teams to understand user needs and translate them into model improvements
Qualifications
Required
- Proven experience with large-scale LLMs and Deep Learning systems
- Strong programming skills in Python/PyTorch; versatility is a plus
- Experience with post-training techniques and reinforcement learning
- Self-starter with a willingness to take ownership of tasks
- Passion for tackling challenging problems
- Minimum 2-6 years of experience on relevant projects (depending on seniority level)
Nice-to-have
- PhD in Machine Learning, AI, Systems, or related areas
- Experience in post-training LLMs with SFT/DPO/GRPO
- C++/CUDA programming skills
- Experience building LLM training frameworks
- Academic publications and research impact
- Experience with agent systems and multi-step reasoning
- Background in personalization and preference learning
Skills
PyTorch, Python, LLMs, Reinforcement Learning, Sft, Dpo, Grpo, Megatron, CUDA, C++
Similar jobs
AI Research jobs
Perplexity
San Francisco, CABuild and evolve the agent harness powering Perplexity’s flagship answer experience, improving orchestration, context management, performance, reliability, observability, and evaluation. The role requires strong software engineering skills, Python proficiency, and experience shipping large-scale AI systems.
Fireworks AI
San Mateo, CAConduct foundational research on LLMs and multimodal systems, designing architectures and training methods and helping move prototypes into production. The role targets PhD researchers graduating by December 2026 with strong machine-learning research and programming experience.
OpenAI
San Francisco, CA
Researcher developing and publishing mechanistic interpretability techniques, building infrastructure to study model internals, and guiding alignment-focused research. Requires research experience in machine learning or a related field, strong engineering skills, and proficiency in Python or similar languages.
Applied Intuition
Sunnyvale, CA
Research Engineer developing and deploying machine-learning algorithms for autonomous driving and robotics systems. The role targets recent MS or PhD graduates with experience in areas such as foundation models, diffusion policies, reinforcement learning, computer vision, and robotics.
GrayMatter Robotics
Los Angeles, CA
Conduct research and develop foundation models for robotic manipulation and high-precision manufacturing, taking projects from data curation through deployment on industrial robots. The role requires current PhD study, strong Python and deep learning expertise, robotics simulation experience, and research in foundation models.