Senior ML Research Engineer - Search & Retrieval

Perplexity; Berlin (NH); On-site; USD 140,000 - 200,000; Full time

Job summary

Perplexity is seeking an experienced Machine Learning Research Engineer to advance the next generation of search technologies, with a focus on retrieval and ranking. You will design, train, and optimize large-scale models using PyTorch, employ distributed training with DeepSpeed/FSDP, and build core components of the search platform.

Collaborate with Data, Infrastructure, and Product teams to ensure fast, high-quality delivery and progress in representation learning for multilingual and

Qualifications

• Deep understanding of search and retrieval systems.
• Experience with large-scale search or recommender systems.
• Strong proficiency with PyTorch and distributed training.
• Expertise in representation learning and multilingual/multimodal modeling.
• Strong publication record in AI/ML conferences.

Responsibilities

• Relentlessly push search quality forward through models, data, tools, or leverage available
• Architect and build core components of the search platform and model stack
• Design, train, and optimize large-scale deep learning models using PyTorch with distributed training and hardware acceleration
• Conduct advanced research in representation learning, including contrastive learning, multilingual, and multimodal modeling
• Deploy models in a scalable and performant way
• Build and optimize RAG pipelines for grounding and answer generation
• Collaborate with Data, AI, Infrastructure, and Product teams to ensure fast and high-quality delivery

Skills

Search systems

Retrieval models

PyTorch

Distributed training

Representation learning

Multimodal modeling

Publications

Ownership

3-5 years experience

Tools

DeepSpeed

FSDP

PyTorch

Job description

Perplexity is seeking an experienced Machine Learning Research Engineer to advance the next generation of search technologies, with a focus on retrieval and ranking. You will design, train, and optimize large-scale models using PyTorch, employ distributed training with DeepSpeed/FSDP, and build core components of the search platform.

Collaborate with Data, Infrastructure, and Product teams to ensure fast, high-quality delivery and progress in representation learning for multilingual and