About the job
The Allen Institute is seeking a Research Engineer to join the FlexOlmo team in Seattle, WA. This position focuses on advancing large language model architectures, especially in Mixture-of-Experts (MoE), long-context language models (LCLMs), and flexible data utilization. Work will be based at our Seattle offices. For details about on-site expectations, please reach out to the recruiter.
About the FlexOlmo Team
FlexOlmo designs new model architectures and training methods to help models use data more effectively. The team explores improved training strategies, inference-time conditioning, and retrieval, broadening the types of data models can leverage and ultimately raising performance. FlexOlmo also develops scientific approaches for evaluating and understanding these systems. The team’s work includes impactful research and open-source tools that support NLP research worldwide. The initial release in July 2025 introduced a significant Mixture-of-Experts architecture, with future projects targeting further innovation in AI.
About the Allen Institute
The Allen Institute for AI (Ai2) is a non-profit focused on foundational AI research and innovation. The organization creates large-scale open models, datasets, and artifacts such as OLMo, Tulu, Asta, and OlmoEarth. Teams at Ai2 bring together leading scientific and engineering talent to pursue open AI, moving quickly from idea to action through collaboration.
What You Will Do
- Contribute to the development of next-generation language model architectures, with an emphasis on Mixture-of-Experts and long-context models
- Apply and extend methods for flexible data utilization during training and inference
- Translate high-level research goals into concrete implementation steps and methodologies
- Work collaboratively within a team of researchers and engineers
- Present findings and explain complex technical concepts clearly
- Help create open-source tools that support the NLP research community
What We’re Looking For
- Hands-on engineering experience in deep learning and natural language processing
- Strong understanding of language models and transformer architectures, especially Mixture-of-Experts
- Ability to work independently and collaboratively
- Experience translating research objectives into actionable steps
- Clear communication skills for technical and non-technical audiences
- Interest in advancing open AI and contributing to impactful research
Compensation
Base salary range: $128,880 to $193,320. Compensation includes generous bonus plans.
