Pre-training Research Engineer

Sciforium•San Francisco, CA

7d•$165,000 - $225,000

About The Position

Sciforium is an AI infrastructure company developing next-generation multimodal AI models and a proprietary, high-efficiency serving platform. Backed by multi-million-dollar funding and direct sponsorship from AMD with hands-on support from AMD engineers the team is scaling rapidly to build the full stack powering frontier AI models and real-time applications. As a Pre-training Research Engineer, you’ll focus on model implementation, pertaining and scaling, and improving the quality of our byte-native and multimodal foundation models. You’ll build and iterate quickly on research ideas, contribute production-grade training code and infrastructure, and help deliver high-quality base models that can serve real-world use cases at scale.

Requirements

5+ years of experience in machine learning research or engineering, with a proven track record of developing and pre-training large language or multimodal foundation models.
Strong general software engineering skills, with the ability to write robust and performant training code.
Solid understanding of deep learning fundamentals and modern pre-training methods and literature.
Ability to quickly implement research ideas and evaluate them using clear baselines, ablations, metrics, and analysis.
Hands-on experience running training workloads in GPU-based environments, with familiarity with distributed training.
MS in Computer Science, Machine Learning, Artificial Intelligence, Mathematics, or a related field.

Nice To Haves

PhD in Computer Science, Machine Learning, Artificial Intelligence, Mathematics, or a related field.
Extensive experience with the JAX, Flax, and XLA stack.
Experience with multi-node pre-training using systems such as FSDP, ZeRO, or Megatron.
Experience developing training recipes, ablations, or scaling experiments.
Experience owning end-to-end training and evaluation pipelines with monitoring and reproducibility.

Responsibilities

Train large byte-native and multimodal foundation models across massive, heterogeneous corpora.
Implement and evaluate new model architectures, training objectives, and optimization methods.
Develop stable pre-training recipes and run scaling experiments for novel architectures.
Conduct ablations and analyze training dynamics, model behavior, and base-model quality.
Work with data and distributed training engineers to improve training efficiency, reliability, and scalability.