AI Researcher (Training Optimization)

TL;DR

AI Researcher (Training Optimization): Developing novel techniques to reduce training cost, accelerate convergence, and improve model quality for large-scale models with an accent on optimization algorithms, schedulers, normalization, and curriculum strategies. Focus on designing experiments, analyzing training dynamics, implementing mixed-precision training, gradient noise reduction, and authoring research papers.

Remote (world)

Company

hirify.global builds efficient, scalable large-scale model training infrastructure.

What you will do

• Design and evaluate training optimization techniques for large models including optimizers, schedulers, normalization, and curriculum strategies.
• Improve training efficiency and stability across long runs and large datasets.
• Research and implement methods like mixed-precision training, gradient noise reduction, scaling laws, and training-time regularization.
• Run large-scale experiments, analyze results, and translate findings into improvements.
• Author research papers, technical reports, or blog posts.
• Collaborate with infrastructure and inference teams to align training with production performance.

Requirements

• 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).
• Solid understanding of optimization theory, backpropagation, gradient flow, training stability, and distributed/large-batch training.
• Proficiency in Python and modern ML frameworks (PyTorch preferred).
• 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.

Culture & Benefits

• Real influence over core model training decisions.
• Freedom to pursue and publish novel research.
• Direct access to large-scale experiments and production constraints.
• Small, senior team valuing deep thinking and thoughtful shipping.