AI Researcher — Training Optimization
J
JobgetherArtificial Intelligence
United StatesFull-TimeSenior
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Required Skills
- PythonMachine LearningPyTorchDeep LearningLLM
Requirements
- Strong background in machine learning research, with expertise in training dynamics, optimization, and large-scale model training.
- Demonstrated experience training large neural networks, such as LLMs or multimodal models.
- Publication experience at venues such as NeurIPS, ICML, ICLR, ACL, EMNLP, or COLM.
- Strong understanding of optimization theory, backpropagation, gradient flow, and distributed training.
- Strong proficiency in Python.
- Experience with PyTorch.
- Ability to independently formulate research questions, design experiments, and interpret complex datasets.
- Experience with non-standard architectures is a plus.
- Experience optimizing large-scale GPU training (FSDP, ZeRO) is desirable.
Responsibilities
- Design, implement, and evaluate novel training optimization techniques for large-scale neural networks.
- Investigate approaches for improving training efficiency, stability, convergence speed, and overall model quality.
- Research and implement techniques involving optimizer innovations, mixed-precision training, and gradient noise reduction.
- Design and execute large-scale experiments and translate findings into actionable improvements.
- Author or co-author research papers, technical reports, and blog posts.
- Collaborate with infrastructure and inference engineering teams to connect research to production performance.
View Full Description & ApplyYou'll be redirected to the employer's site
