AI Researcher — Training Optimization

The role

You will design and evaluate training optimization techniques for large models. You will work at the intersection of research and systems to improve efficiency, stability, and scalability.

What you will do

• Design and evaluate training optimization techniques for large models
• Improve training efficiency and stability across long runs and large datasets
• Research and implement optimizer and scheduler innovations
• Research and implement mixed-precision, low-precision, and memory-efficient training
• Research and implement gradient noise reduction, scaling laws, and convergence analysis
• Research and implement training-time regularization and robustness techniques
• Run large-scale experiments, analyze results, and translate findings into actionable improvements
• Author or co-author research papers, technical reports, or blog posts
• Collaborate with infrastructure and inference teams

Required qualifications

• Strong background in machine learning research, with emphasis on training dynamics and optimization
• Experience training large neural networks (LLMs, multimodal models, or large sequence models)
• Publication experience in ML venues (e.g. NeurIPS, ICML, ICLR, ACL, EMNLP, COLM, arXiv) or equivalent high-quality open research
• Solid understanding of optimization theory and practice
• Solid understanding of backpropagation, gradient flow, and training stability
• Solid understanding of distributed and large-batch training
• Proficiency in Python and modern ML frameworks
• Ability to independently design experiments and reason from data

Nice to have

• Experience with non-standard architectures (e.g. RNN variants, long-context models, hybrid systems)
• Experience optimizing training on GPUs at scale (FSDP, ZeRO, custom kernels)
• Contributions to open-source ML or research codebases
• Comfort operating in fast-moving, ambiguous startup environments

Benefits

• Real influence over core model training decisions
• Freedom to pursue and publish novel research
• Direct access to large-scale experiments and real production constraints

About the company

featherlessai is a small, senior team that values thinking deeply and shipping thoughtfully.