Biohub is the first large-scale initiative bringing frontier AI models, massive compute, and frontier experimental capabilities under one roof. We're building a general-purpose system to accelerate scientific discovery, integrating frontier AI models, biological foundation models, and lab capabilities, with the ultimate goal of curing disease. Our technology powers scientists around the world, translating AI capabilities into tools that accelerate research everywhere.
The Team
Biohub is the first large-scale initiative bringing frontier AI models, massive compute, and frontier experimental capabilities under one roof. We're building a general-purpose system to accelerate scientific discovery, integrating frontier AI models, biological foundation models, and lab capabilities, with the ultimate goal of curing disease. Our technology powers scientists around the world, translating AI capabilities into tools that accelerate research everywhere.
Biohub operates one of the largest AI compute clusters dedicated to biology, spanning three frontier research institutes with some of the world's leading biologists. We're not a startup trying to find product-market fit, and we're not a pharma company optimizing a pipeline. We're building frontier AI for fundamental science, as open science, at a scale no one else is doing. This is a unique moment for scientific acceleration. The problems are among the hardest and most impactful problems you can choose to work on, and we move at a pace that meets this moment.
Our research spans:
Frontier molecular modeling, from protein language models (e.g., ESM) to structure prediction (e.g., ESMFold) and beyond.
Scaled biological foundation models trained on some of the largest GPU clusters dedicated to science
Imaging foundation models trained across the world's largest microscopy datasets
Reasoning and agentic systems that connect frontier LLMs with biological foundation models
Mechanistic interpretability of biological foundation models: extracting new biological knowledge directly from model weights
Scientific data at unprecedented scale: AI systems to collect, curate, and learn from some of the richest biological datasets ever assem bled
The Opportunity
As an ML Engineer, you'll join some of the strongest infrastructure engineers in AI, building the systems that connect everything together. The infrastructure problems you solve directly determine what science becomes possible.
What You'll Do
Our team has a wide range of responsibilities:
Work with researchers to train, fine-tune, and evaluate biological foundation models and diagnose training problems.
Help evaluate and curate training and validation datasets at Petabytes scale.
Design and create libraries and abstractions that researchers and engineers build on.
Optimize GPU-native data loading for large tensor workloads and resolve hardware utilization bottlenecks across backends.
Build and maintain reliable distributed pre and post training systems.
Support production deployment, with tracking of model and dataset artifacts throughout the ML lifecycle.
Improve model and cluster monitoring, unit and integration testing, and continuous integration.
Work with partner researchers and engineers to deploy models and supporting infrastructure in external environments.
What You'll Bring
Significant experience in at least one of the following areas:
5+ years building ML systems or distributed infrastructure at scale.
Strong Python, PyTorch, and CUDA skills, including custom training loops, with a solid understanding of model architectures, optimization, and evaluation.
Experience with large scale distributed systems, multi-GPU training, checkpointing, and failure recovery.
Ability to diagnose performance problems across GPUs, memory, networking, storage, and data loading.
Experience building large-scale data pipelines and shared research tools; familiarity with scientific formats such as Zarr or HDF5.
Experience deploying and monitoring models, tracking…