Job SummaryGDM is building an agentic, prescriptive breeding platform. The Computational Breeding & Molecular Omics Lead owns GDM's molecular omics science: integrating the molecular -omes (transcriptomics, genomics, epigenomics, metabolomics and related layers), driving next-generation genomics discovery, and designing GDM's next-generation marker panel - turning molecular signal into earlier, more prescriptive breeding decisions.
This role brings the molecular and computational-biology expertise GDM's breeding programs cannot generate on their own, and translates it into breeding value - delivering model-ready molecular features to the prediction stack and building the molecular applications that produce them. The role sits alongside the platform, prediction and agent-development functions: it owns the molecular science; the orchestration platform and the master AI agent are owned elsewhere.
Duties and ResponsibilitiesMulti-molecular-omics integration & next-generation genomics- Integrate GDM's molecular -omes - transcriptomics, genomics, epigenomics, metabolomics and related layers - into coherent, selection-relevant molecular signal, rather than isolated single-omic analyses.
- Lead next-generation genomics discovery: evaluate and bring in the next wave of genotyping/sequencing and molecular approaches that raise resolution and lower cost.
- Design and own GDM's next-generation marker panel alongside the Biotech Lab - what it captures, at what density and cost, and how it improves genotyping across the pipeline; the prediction function validates the panel against accuracy.
- Keep improving GDM's genotyping and molecular capability.
Molecular signal for breeding decisions & Speed Breeding- Identify which molecular signals - from WGS, transcriptomics or epigenomics - predict breeding value before phenotypic data is available, enabling earlier advancement decisions.
- Make Speed Breeding prescriptive: define the molecular markers and omics signals that allow selection in early generations, before field phenotype is measurable - the molecular contribution to GDM's Accelerated Cycle pillar.
- Define which molecular decision points can be automated and at what stage of the breeding cycle.
Delivery to prediction & molecular applications- Deliver all molecular data and features to the prediction function in model-ready format; if they improve accuracy they enter production, if not the role adjusts.
- Coordinate with Traits Co to generate molecular data to required specification (format, quality, developmental stage) so it is model-ready from the moment it is generated.
- Build production-grade molecular applications and bioinformatics pipelines - documented, version-controlled, API-connectable - not standalone analysis scripts; define each application's inputs and outputs so the master agent can call them.
- Molecular agents may be built in this role, but the master orchestration agent is owned by the AI Agent Engineer; this role owns the molecular logic, the platform decides integration.
Connecting the pieces - making disparate data sources usable- Act as a connector, cross-functional resource: take heterogeneous, disconnected data sources - molecular, genotypic, pipeline, phenotypic and external - and identify how they relate so they can be used together.
- Thread sources together into a coherent picture: map keys, resolve identifiers, reconcile formats and surface the relationships that let molecular data join genotype, phenotype and environment downstream.
Feeds and is fed by adjacent functions- Data Analytics - feeds: model-ready molecular features and new-panel signal into the data platform. Fed back: data requirements, format/quality standards, and the pipelines and infrastructure on which the molecular applications run.
- Breeding Models (Prediction) - feeds: molecular features for the prediction models. Fed back: accuracy feedback - which features and panels improve prediction and which do not - which prioritizes discovery and panel design.
- AI Agent Engineer - feeds: molecular applications with clearly defined inputs and outputs (API contracts) so they can be orchestrated. Fed back: orchestration - the agent calls the molecular applications in the right sequence within the decision workflow.
- Traits Co - feeds: molecular data specifications (what to generate, at what quality and stage). Fed back: molecular data generated to specification.
Required Skills and Abilities- Strong molecular omics and computational-biology background across multiple layers - genomics plus at least one of transcriptomics / epigenomics / other -omes - applied to genetics or breeding.
- Demonstrated work in genomics innovation - marker/panel design, next-generation genotyping or sequencing, or integrating multiple molecular data types into usable signal.
- Production-grade computational skills (Python, R and/or HPC) - building pipelines and applications others use, delivered documented, reproducible, version-controlled and API-connectable.
- Ability to translate molecular findings into breeding implications - an earlier, more confident selection decision, not biological interest for its own sake.
- Preferred (strong plus): experience building AI agents or automated pipelines for genomic processes; Speed Breeding / early-generation molecular selection; row crops (soybean, corn, wheat, sunflower); commercial or AgTech environments where scientific output drives business decisions.
Education and Experience- PhD in computational biology, plant genomics, bioinformatics, plant genetics, or a closely related field.
- Track record delivering working software/tools that others use (not only analysis scripts), applied to plant genomics / molecular biology problems.
- Not required: end-to-end platform architecture or orchestration; ownership of the master orchestration agent; multimodal prediction modelling (genome+phenome+envirome); breeding-scheme design or simulation; prior commercial breeding-industry experience.
Physical Requirements- Prolonged periods of working at a computer and using standard office / compute equipment.
- Frequent national and international travel to GDM research sites and partner facilities is required.
