Logo
Umbilical Life

Machine Learning Research Engineer

Umbilical Life, Boston

Save Job

Overview

I am working with a leading Tech Bio company in Boston, looking for a Senior (Senior/Principal/Staff Scientist) Machine Learning Research Engineer to lead the development of their Biological AI Model.

The candidate should be local to Boston on a weekly basis.

Key Responsibilities

  • Design and implement core AI/ML models for simulating cellular systems using multi-omics and single-cell data.
  • Develop novel architectures e.g. Graph Neural Networks, Causal Inference, Transformers, diffusion models, VAE etc. tailorable to biological complexity.
  • Contribute to the strategic direction of modeling efforts , helping define what to build, why, and how.
  • Lead model design from prototyping to production
  • Guide internal thinking around biological networks, perturbation models , and high-dimensional cellular data.
  • Support cross-functional collaboration and help define a scalable modeling stack and modeling best practices across the company.

Qualifications

  • MS or PhD in Computer Science, Physics, Applied Math, or similar , with a strong focus on AI/ML.
  • Strong track record in research outputs on single cell, AI method development.
  • Expertise in building models using GNNs, VAEs, Transformers , reinforcement learning or other deep learning approaches.
  • Strong proficiency in Python and deep learning frameworks such as PyTorch, TensorFlow or JAX.
  • Exposure to single-cell data (e.g., scRNA-seq, spatial omics)
  • Strong ability to abstract and model complex biological processes from a data/physics/ML perspective.
  • Experience with scaling models across biological levels - from individual cells to tissues and whole organisms, is a strong plus, given the complexity of multi-scale integration .
  • Experience working on noisy, high-dimensional, multi-modal biological data sets .
  • Curious, collaborative, and comfortable in fast-moving, exploratory R&D environments.
  • Previous experience with Virtual Cell Models is a plus

#J-18808-Ljbffr