ML Research Engineer (Data Engineering)

Metamorphic•Palo Alto, CA

About The Position

Metamorphic is developing new approaches to intelligence by combining machine learning with large-scale experimental neuroscience, informed by the principles that make the brain efficient, flexible, and robust. We are building foundation models trained on rich, continuous neural data — a high-resolution model of the brain at a scale never before possible. Our founding team spans machine learning, neuroscience, and neurotechnology, with prior work including the MICrONS project, Neuropixels, and the Enigma project, as well as foundational scientific contributions in learning, neural computation, and generative modeling. Our work sits at the frontier of AI research, and we believe the highest-impact discoveries will come from researchers and engineers working as a single, tightly collaborative team. The name Metamorphic reflects our belief that the next advances in intelligence will come from a change in form, beyond scale — from artificial to natural intelligence. About the Role We are hiring Research Engineers to sit at the boundary of research and systems engineering. Our multimodal data spans trillions of tokens of video alongside rich neural and behavioral recordings, making this one of the most demanding dataloading challenges in frontier AI. You will own the systems that turn this large, heterogeneous data into training-ready multimodal streams for foundation model training and evaluation at scale. This means designing and building a state-of-the-art end-to-end dataloading stack: data formatting, preprocessing, filtering, sharding, caching, and streaming. You will build runtime interfaces that deliver data to distributed training jobs across GPU clusters with high throughput, reliability, and full observability. You'll have substantial autonomy to shape foundational technical decisions on a small, high-impact team. You'll thrive in this role if you: Are excited about working in a fast-paced, production-focused research lab that often requires switching between many hats Have significant software engineering experience and can move quickly without sacrificing rigor Are able to balance research goals with practical engineering constraints Enjoy pair programming and deeply collaborative work Are eager to learn more about machine learning research in a novel scientific domain Are enthusiastic to work at an organization that functions as a single, cohesive team pursuing large-scale AI research Have ambitious goals for AI progress and are excited to create the best outcomes over the long term

Requirements

Bachelor's degree or equivalent experience in Computer Science, Machine Learning, Computational Neuroscience, or a related field
Strong software engineering skills in Python and familiarity with PyTorch
Experience building high-throughput data pipelines or dataloading systems for large-scale distributed ML training
Experience working with and building systems for complex, multimodal time-series data
Experience with video processing at scale: decoding, transcoding, I/O optimization for large video corpora
Hands-on experience profiling and benchmarking data systems on metrics such as throughput, IOPS, GPU utilization, and memory usage

Nice To Haves

Familiarity with multi-modal transformer architectures
Experience with distributed training environments and deep understanding of sharding models and data
Experience with ML workflow orchestrators (e.g. Prefect, Dagster, Airflow).
Experience with containerization, and scaling container orchestration (e.g. via Docker, Kubernetes)
Experience with performance-critical compiled or systems languages (e.g. Rust, C++, CUDA)
Proficiency with MLOps platforms for experiment tracking and reproducibility (e.g. MLflow, W&B)
Background in scientific computing, computational neuroscience, life sciences, or ML-adjacent research environments

Responsibilities

Own the systems that turn large, heterogeneous data into training-ready multimodal streams for foundation model training and evaluation at scale.
Design and build a state-of-the-art end-to-end dataloading stack: data formatting, preprocessing, filtering, sharding, caching, and streaming.
Build runtime interfaces that deliver data to distributed training jobs across GPU clusters with high throughput, reliability, and full observability.