# Staff Research Engineer, Model Efficiency

**Company:** [Cohere](https://hotfix.jobs/companies/cohere)
**Location:** Remote
**Role:** ML Engineering
**Skills:** LLMs, Model Architecture, Moe, Inference Optimization, Gpu Acceleration, PyTorch, JAX, Machine Learning, Software Engineering, Decoding Algorithms
**Posted:** 2025-11-07

> Develops and deploys techniques to enhance LLM inference efficiency, focusing on architecture optimization, decoding algorithms, and GPU acceleration. Requires PhD in ML, expertise in LLM optimization, strong software skills, and top-tier publications.

## Job Description

## Responsibilities
- Develop, prototype, and deploy techniques that improve LLM inference efficiency in production.
- Explore breakthroughs across model execution stack, including model architecture and MoE routing optimization.
- Implement decoding and inference-time algorithm improvements.
- Perform software/hardware co-design for GPU acceleration.
- Optimize performance without compromising model quality.

## Requirements
- PhD in Machine Learning or related field.
- Deep understanding of LLM architecture and optimization under resource constraints.
- Significant experience with model efficiency techniques.
- Strong software engineering skills.
- Experience in fast-paced, high-ambiguity startup environment.
- Publications at top-tier conferences (ICLR, ACL, NeurIPS).
- Passion for mentoring others.

## Nice-to-Haves
- Appetite for working in startups (encouraged even if not perfect fit).

## Similar roles

- [Member of Technical Staff](https://hotfix.jobs/jobs/dd7c8bc4-741a-45d5-91a8-e6b1c4ad67e0) - xAI - Palo Alto, CA - $180k – $440k/yr
- [Staff Software Engineer, AI & App Experience](https://hotfix.jobs/jobs/78128221-8199-41cd-9c63-502d3f61d603) - Chime - New York, NY - $223k – $308k/yr
- [Staff + Senior Software Engineer, Cloud Inference Launch Engineering](https://hotfix.jobs/jobs/18b08a6a-5359-4836-b72e-c641fa718b04) - Anthropic - San Francisco, CA - $320k – $485k/yr
- [Senior Staff Software Engineer, Legal Automation](https://hotfix.jobs/jobs/7870d3e1-ebec-4f7b-9ae5-c62926a06d6e) - Coinbase - Remote - $254k – $299k/yr
- [Staff Machine Learning Engineer, Computer Vision](https://hotfix.jobs/jobs/2969c5c0-8479-43bb-b9b6-f8df93b528d7) - Pinterest - San Francisco, CA - $189k – $390k/yr

**Apply:** https://hotfix.jobs/jobs/1433564e-05ba-488f-9afa-b7a0bba6cce5
**Canonical:** https://hotfix.jobs/jobs/1433564e-05ba-488f-9afa-b7a0bba6cce5