Data Scientist - Innovation - PhD (Irving, TX)

Caris Life Sciences•Irving, TX

1d•Onsite

About The Position

Want to help build AI models for the next generation of cancer diagnostics? The models you build here have direct line-of-sight to translational research and clinical decision-making -- work with the potential to shape how cancer is detected, profiled, and treated. As a Data Scientist on the Innovation Team, you will develop machine learning and deep learning algorithms on molecular sequencing data (WGS, WES, RNA-seq, cfDNA), design analytic pipelines for novel biomarker discovery, and tackle the most challenging problems in liquid biopsy and translational oncology research. The Innovation Team is a small, fast-moving R&D group within Caris Life Sciences, drawing on proprietary clinical research data that no other team in oncology can match. We work closely with bioinformaticians, molecular biologists, and clinical scientists to develop high-impact AI models with the potential to shift the landscape of clinical outcomes. You will have the freedom to lead research projects end-to-end -- from problem framing to deployment -- and to shape the methods that drive Caris' R&D agenda. In your first year, success looks like leading one or two research projects from problem framing through deployment, contributing to a peer-reviewed publication or conference submission, and helping shape methods that inform Caris' diagnostic platform.

Requirements

PhD in Data Science, Bioinformatics, Computational Biology, Genomics, Statistics, Computer Science, Engineering, Biophysics, or a related quantitative or biological field.
PhD recently completed, or up to approximately 2 years of post-doctoral research experience (academic or industry).
Demonstrated work on a cancer biology or translational research problem (PhD thesis chapter, peer-reviewed publication, or postdoc / industry role).
Hands-on experience with molecular sequencing data (e.g., WGS, WES, RNA-seq, cfDNA) including production-grade pipelines and analysis.
Hands-on experience with generative AI -- large language models, foundation models (e.g., genomic or protein language models), or agentic workflows applied to scientific or clinical data.
Proficiency with PyTorch and modern deep learning architectures (transformers, attention mechanisms), with demonstrated application of ML/DL to biological or clinical data.
First-author or co-first-author peer-reviewed publications in machine learning venues (e.g., NeurIPS, ICML, ICLR) or in bioinformatics / computational biology journals.
Strong Python; comfortable in Linux; proficient with git and collaborative workflows.

Nice To Haves

Multi-omics integration experience (genomics, transcriptomics, proteomics, methylation, etc.).
Experience with epigenetics -- DNA methylation analysis, chromatin biology, or related.
Interest in cell-free DNA, liquid biopsy, and next-generation early cancer diagnostics.
Interest in novel algorithm development for biomedical signal extraction in sequencing data.
Proficiency in cloud platforms (AWS EC2, S3, HealthOmics) and containerization (Docker).

Responsibilities

Processing, manipulating, and analyzing large diverse datasets generated from NGS to develop biomarkers for cancer diagnosis, prognosis, and treatment.
Developing novel algorithms for feature extraction and biomarker discovery from molecular sequencing data.
Applying first-principles analysis to translate open research questions into tractable, well-defined problems.
Applying state-of-the-art machine learning and deep learning methods to biological and clinical research questions.
Creating rigorous evaluation frameworks and tracking experiments systematically using tools such as MLflow or Weights & Biases.
Authoring peer-reviewed research publications and presenting findings at scientific conferences.