Machine Learning Researcher, Genomic AI

Bayer•Tulsa, OK

8d•$110,000 - $150,000•Remote

About The Position

We are seeking a Machine Learning Researcher with expertise in machine learning for biological systems, with a particular focus on genomic and multi-omic data modeling. This role is centered on building and deploying state-of-the-art AI models- including large-scale genomic language models and deep representation learning architectures - that extract actionable biological insight from complex molecular datasets. You will develop models that learn the grammar of genomes, predict functional consequences of genetic variation, and connect molecular signatures to whole-organism phenotypes across diverse crop species. Your work will directly enable transformative applications in genomic selection and genome editing target identification, translating sequence-level intelligence into breeding and discovery decisions at global scale. This position is being hired at the entry level. Depending on the candidate's depth of experience and demonstrated research impact, the role may be filled at the Senior Machine Learning Researcher level.

Requirements

PhD in one of the following or closely related fields: Computational Biology / Bioinformatics, Machine Learning / Deep Learning, Genomics / Statistical Genetics, Computer Science (with focus on biological or sequential data), Biostatistics / Quantitative Genetics, Systems Biology, Or another related quantitative discipline with demonstrated application to biological data
Demonstrated research experience building and training deep learning models on biological sequence data or high-dimensional omic datasets.
Proficiency in modern deep learning frameworks (PyTorch, JAX, or TensorFlow) and familiarity with large-scale model training (distributed training, GPU clusters).
Working knowledge of molecular biology fundamentals sufficient to interpret model outputs in biological context (e.g., gene regulation, variant consequence, population genetics).
Strong communication skills and ability to collaborate effectively across disciplines.

Nice To Haves

Hands-on experience developing or fine-tuning genomic language models or biological foundation models (e.g., GPN, PlantCaduceus, Nucleotide Transformer, Evo, Enformer, AlphaGenome or similar large-scale sequence architectures for genomic prediction and functional track prediction).
Experience with transformer architectures, long-context sequence modeling, or attention mechanisms applied to biological sequences.
Familiarity with multi-omic data integration methods (e.g., multi-modal autoencoders, contrastive learning across modalities, graph neural networks on biological networks).
Background in quantitative genetics or genomic prediction (e.g., GBLUP, Bayesian alphabet models, marker-effect estimation) and understanding of breeding program workflows.
Experience with functional genomics data: ATAC-seq, ChIP-seq, Hi-C, single-cell transcriptomics, or CRISPR screen data.
Knowledge of pangenomics, structural variant calling, or comparative genomics across crop species.
Experience with self-supervised, semi-supervised, or transfer learning strategies for data-efficient modeling in biology.
Familiarity with interpretability/explainability methods (attention visualization, in-silico mutagenesis, feature attribution) to derive biological hypotheses from model internals.
Exposure to classical ML approaches (gradient-boosted methods, kernel methods, Gaussian processes) as complementary or baseline tools.
Experience with model deployment in production (MLOps pipelines, containerization, API development, cloud/HPC infrastructure).
Track record of interdisciplinary collaboration with experimental biologists, resulting in validated biological predictions.

Responsibilities

Design, train, and evaluate deep learning models (including large language models, transformers, and representation learning architectures) on diverse omic datasets - whole-genome sequences, gene expression profiles (RNA-seq), epigenomic marks, k-mer spectra, skim-seq, pangenome graphs, and multi-omic integrations.
Develop and fine-tune foundation models for DNA/RNA sequences that capture long-range dependencies, regulatory grammar, and evolutionary conservation to predict variant effects, gene function, and trait associations in crop genomes.
Build predictive models that connect genotype to phenotype across environments, identify high-value editing targets, and rank candidate genetic interventions with biological interpretability and statistical rigor.
Integrate heterogeneous biological data types-including high-resolution genome assemblies, structural variants, gene regulatory networks, protein structure predictions, and phenomic measurements-into unified predictive frameworks.
Work closely with molecular biologists, geneticists, breeders, bioinformaticians, and computational scientists to ground models in biological reality, design informative training data strategies, and validate predictions experimentally.
Partner with engineering and IT teams to operationalize models within genomic selection pipelines, editing nomination workflows, and decision-support platforms used by breeding programs globally.
Advance the state of the art through publications, internal seminars, and engagement with the broader computational biology and AI research community.
Communicate complex modeling results to diverse audiences, prepare technical reports, and build organizational confidence in AI-driven biological discovery.