Lead research and applied ML initiatives focused on building and scaling genomic foundation models, generative design of CRISPR enzymes for gene editing, and high-throughput in silico target screening supporting biological discovery, predictive design, and product development.
Key contributions:
• Led a small, cross-functional team in the design and training of a biologically-informed foundation model for plant genomics, improving predictive performance of gene regulation and enabling automated screening and design workflows across multiple research programs.
• Architected and deployed an internal ML platform integrating protein structure prediction, generative modeling, and workflow orchestration to support high-throughput biological screening in a secure environment.
• Built scalable protein-protein interaction screening and prioritization pipelines that significantly reduced experimental load and accelerated target discovery cycles.
• Developed cross-species transfer learning approaches to integrate and enrich heterogeneous biological datasets, improving inference quality in data-sparse regimes.
• Automated quantitative phenotyping workflows using computer vision and graph-based analysis to transform large collections of noisy images into interpretable features.
