Research Scientist

Oumi Research Scientist

The Research Scientist will be an integral part of Oumi's research team, focusing on advancing the state-of-the-art in large language models (LLMs), vision language models (VLMs), and related technologies. This role involves conducting cutting-edge research, contributing to open-source projects, and collaborating with other researchers and engineers. Researchers at Oumi will work on various aspects of LLM/VLM development including training, evaluation, data curation, and benchmark development.

What you'll do:

• Conduct research on training and evaluating new Large language models (LLMs), Vision Language Models (VLMs), and other AI models. This includes exploring new architectures, training techniques, and optimization methods.
• Develop methodologies for curating high-quality datasets for training and evaluating LLMs. This may involve data synthesis and other novel techniques.
• Develop evaluation benchmarks to measure the performance of LLMs across various tasks and domains.
• Design and conduct experiments to validate research hypotheses and improve model performance.
• Contribute to the Oumi open-source platform, models and projects, and other relevant tools and libraries.
• Collaborate with other researchers, engineers, and the broader community to advance the field of open-source AI.
• Publish research findings in leading conferences and journals.
• Evaluate existing models and identify areas of improvement.
• Work with various models, including text and multimodal models, and both open and closed models.
• Focus on the research that matters by skipping the plumbing and moving straight to research, building on the work of others and contributing back.

What you'll bring:

• A Ph.D. or MSc. in computer science, machine learning, artificial intelligence, or a related field is preferred. Candidates with a strong publication record, or equivalent industry experience will be considered.
• Demonstrated experience in conducting original research in machine learning, with a strong publication record in top-tier conferences or journals.
• Deep understanding of machine learning and deep learning concepts, with specific knowledge of large language models (LLMs) and/or vision language models (VLMs).
• Strong programming skills in Python and experience using deep learning frameworks (e.g. PyTorch).
• Familiarity with open-source projects and a passion for contributing to the open-source community.
• A self-starter who can work independently and take ownership of initiatives.
• Share Oumi's values:
  Beneficial for all, Customer-obsessed, Radical Ownership, Exceptional Teammates, Science-grounded.

Benefits:

• Equity in a high-growth startup
• Comprehensive health, dental and vision insurance
• 21 days PTO
• Regular team offsites and events

Location:

• Hybrid work environment, with offices in:
• San Mateo, CA
• Seattle, WA
• New York, NY