Role overview
Ohalo™ aims to accelerate evolution to unlock nature's potential. Founded in 2019, Ohalo develops novel breeding systems and improved plant varieties that help farmers grow more food with fewer natural resources, increasing the yield, resiliency, and genetic diversity of crops to sustainably feed our population. Ohalo's breakthrough technology, Boosted Breeding™, will usher in a new era of improved productivity to radically transform global agriculture. For more information, visit www.ohalo.com.
The anticipated pay range for this role is $170,000 - $220,000 per year for our San Francisco, CA location, though salary will be based on a variety of factors including, but not limited to, experience, skills, education, and location.
Notes: If you previously applied for a job at Ohalo Genetics, we encourage you to restate your interest in the position by submitting your application.
No visa sponsorship is available for this position at this time.
No recruiters, please.
What you'll work on
- Design, build, and maintain scalable ML pipelines on GCP (or the best-fit cloud) using Python, BigQuery/Spark, Kubernetes, and CI/CD best practices.
- Mentor & grow a small team—provide technical guidance, establish code-review norms, and cultivate a culture of rapid, well-engineered experimentation.
- Own model-ops lifecycle: automated testing, containerized deployment, continuous monitoring, and A/B evaluation against breeding KPIs.
- Collaborate cross-functionally with plant scientists, data engineers, and the automation group to ingest high-throughput phenotyping data and close feedback loops.
- Establish MLOps Excellence: Build robust infrastructure for model versioning, data lineage, and automated retraining; implement "champion-challenger" deployment patterns to safely promote research models to production while ensuring full auditability of breeding decisions.
What we're looking for
- Advanced Image Techniques – Proficiency in leveraging Foundation Models (e.g., SAM, CLIP) and Self-Supervised Learning to automate high-throughput feature extraction, converting unstructured imagery into high-dimensional embeddings for advanced statistical analysis.