Research Scientist, Post-Training

Summary

Join Together AI's Model Shaping team as a Research Scientist in Post-Training to advance methods for making foundation models more useful and reliable. You will analyze limitations of current post-training approaches, design new techniques for model adaptation, and create benchmarks for measuring progress. After experimenting and evaluating your ideas, you will present findings at leading ML/NLP conferences and collaborate to integrate improvements into Together's platform. The role involves defining and driving the research agenda around foundation model training efficiency and performance. You will study recent AI research, conduct experiments, and share findings internally and externally. Collaboration with Machine Learning Engineers to integrate advanced methods into the platform is also expected. Together AI offers competitive compensation, startup equity, health insurance, and remote work flexibility.

Requirements

• Can autonomously design, implement, and validate your research ideas
• Skilled at writing high-quality and efficient code in Python and PyTorch
• Have first-author publications at leading conferences on ML or NLP (ICLR, ICML, NeurIPS, ACL, EMNLP)
• Are a strong communicator, ready to both discuss your research plans with other scientists and explain them to broader audience
• Follow the latest advances in relevant subfields of AI
• Are passionate about seeing your research create real-world impact through Together AI's services and willing to work hands-on with production systems to achieve it

Responsibilities

• Define and drive the research agenda around efficiency and performance of foundation model training
• Study recent results from the broader AI research community, analyzing their relevance to the team’s research directions and ongoing projects
• Conduct experiments to empirically validate your hypotheses and compare the outcomes with relevant related work
• Share your findings both internally and externally (e.g., at top-tier conferences on ML and NLP)
• Partner with Machine Learning Engineers to integrate advanced methods into Together’s Model Shaping platform

Preferred Qualifications

• Reinforcement learning of language models
• Curation of pre-training or post-training datasets and benchmarks
• Robust evaluation of foundation models
• Running large-scale experiments on GPU clusters

Benefits

• Health insurance
• Competitive compensation
• Startup equity
• Remote work