Staff Research Engineer, Model Efficiency

Who are we?

Role Overview

Large Language Models (LLMs) continue to push the boundaries of what AI systems can do — but inference is still the bottleneck. The Model Efficiency team is responsible for pushing the limits of LLM inference efficiency across our foundation models. We explore and ship breakthroughs across the model execution stack, including:

• model architecture and MoE routing optimization
• decoding and inference-time algorithm improvements
• software/hardware co-design for GPU acceleration
• performance optimization without compromising model quality

Please Note: We have offices in Toronto, Montreal, San Francisco, New York, Paris, Seoul and London. We embrace a remote-friendly environment, and as part of this approach, we strategically distribute teams based on interests, expertise, and time zones to promote collaboration and flexibility. You'll find the Model Efficiency team concentrated in the EST and PST time zones, these are our preferred locations.

Key

Responsibilities

As a Staff Research Engineer, you will develop, prototype, and deploy techniques that materially improve how fast and efficiently our models run in production.

Qualifications

• Have a PhD in Machine Learning or a related field
• Understand LLM architecture, and how to optimize LLM inference given resource constraints
• Have significant experience with one or more techniques that enhance model efficiency
• Strong software engineering skills
• An appetite to work in a fast-paced high-ambiguity start-up environment
• Publications at top-tier conferences and venues (ICLR, ACL, NeurIPS)
• Passion to mentor others

Working Location

This role can be based remotely or from one of our office locations listed on the job description - there is no minimum in-office qualification requirement. We care most about hiring exceptional people regardless of locations, though please check the location listed on the posting for guidance around the core time zone or working hours alignment expected for the role.