Software Engineer, ML Infrastructure
Core
Own the data and evaluation of ML infrastructure for Arc's virtual cell models, including the Model Performance Dashboard and the pipeline turning single-cell data into training sets.
Role type
Senior IC ML Infrastructure Engineer
Builds
Model Performance Dashboard, data-to-training pipeline, lab-in-the-loop infrastructure, validation frameworks for agent-written software
Domain
Artificial Intelligence / Computational Biology / Single-cell data
Deliverable
production ML models
Required skills
Python, PyTorch, distributed training (DDP/FSDP/NCCL/SLURM), GPU utilization optimization, data pipeline design, AI agents, validation frameworks, stakeholder management
Preferred skills
single-cell data (AnnData/h5ad), model evaluation harnesses, multi-node distributed training leadership
Technologies
Python, PyTorch, AnnData, h5ad, DDP, FSDP, NCCL, SLURM
Responsibilities
Design and build evaluation tools for model benchmarking; redesign the single-cell data pipeline to use AI agents for curation; build validation frameworks for agent-written code; optimize multi-node training throughput and GPU utilization; own the CI system for model training and evaluation.
Seniority
Senior, hands-on IC
