Hark

Member of Technical Staff, Pretraining

Hark
Apply
10 days ago
San Jose, CA, USASenior

Base Salary

$180k - $450k/yr

Responsibilities

  • Drive research and development in large-scale LLM and multimodal pretraining.
  • Develop data pipelines for curation, filtering, deduplication, and synthetic data generation.
  • Design and implement distributed training strategies, scaling approaches, and optimization techniques.
  • Build and improve pretraining infrastructure, training systems, data pipelines, and compute efficiency.
  • Develop evaluation frameworks and internal benchmarks for pretraining progress and model capability.
  • Collaborate with research and engineering teams to advance foundation model performance and scalability.

Requirements

  • Demonstrated success improving large-scale neural network performance through pretraining data, modeling, or training-system advances.
  • Strong experience with large-scale distributed training frameworks such as Megatron or DeepSpeed.
  • Deep understanding of LLM or multimodal pretraining, data pipelines, scaling behavior, and optimization.
  • Experience with data-driven experimentation, systematic analysis, and debugging at scale.
  • Experience with large-scale training infrastructure and high-performance computing systems.
  • Ability to take strong ownership in a fast-paced, research-driven environment.
  • Experience with multimodal pretraining across text, audio, or vision is preferred.

Categories

AI ResearchML Engineering
Hark

About Hark

1-10 employees
Contact me