A decade of leading computational biology.

Technical lead for single-cell multi-omics in oncology and immunology — shaping computational strategy from discovery through product development.

• Lead end-to-end single-cell multi-omics programs (transcriptomics, proteomics, spatial) that directly inform target selection, mechanism-of-action narratives, and translational strategy for oncology and immunology assets.
• Set the analytical direction across multiple parallel projects — defining standards, reviewing methodology, and aligning computational deliverables with clinical and discovery stakeholders.
• Mentor junior scientists and manage external consultants; grow team capability through code review, design discussions, and structured onboarding.
• Partner with wet-lab, clinical, and CMC teams to translate computational insights into experimental designs and go/no-go decisions on assets in development.

Built the team's production analysis infrastructure and led single-cell projects across clinical and manufacturing programs.

• Architected and shipped automated high-throughput Nextflow/Python pipelines for scRNA-seq, CITE-seq, and multi-modal analyses — turning week-long manual workflows into reproducible runs and enabling consistent decision-making across programs.
• Led single-cell analyses underpinning biomarker discovery for a clinical-stage cell therapy and process characterization for CMC, contributing directly to regulatory-grade documentation.
• Established team conventions for reproducibility (containerization, version-controlled pipelines, QC standards) adopted as the default for new projects.
• Acted as the go-to technical reviewer for colleagues' analyses, raising the bar on rigor and reproducibility across the group.

PhD developing and scaling open-source methods for spatial transcriptomics, adopted by labs and industry worldwide.

• Designed and led Optocoder, a machine-learning pipeline for decoding barcoded transcripts from imaging-based spatial transcriptomics — published in NAR Genomics & Bioinformatics and adopted by external labs.
• Re-architected novoSpaRc (optimal-transport reconstruction of tissue architecture from scRNA-seq) for scalability, unlocking application to whole-tissue datasets; co-authored Nature Protocols paper.
• Co-inventor on a US patent for 3D spatial gene-expression reconstruction.
• Owned the full research lifecycle: problem framing, method development, benchmarking, open-source release, documentation, and supporting external users.

Independent collaboration applying protein LLMs to agricultural biotech discovery.

• Led an independent short-term project applying protein language models (ProtBERT) to discover novel anti-insecticidal proteins from raw sequence data — combining transfer learning with biological priors to guide downstream experimental validation.

Delivered a production-ready deep-learning system for pharmaceutical QC.

• Developed and shipped deep-learning models for automated particle detection and classification from flow-microscopy images used in biopharmaceutical QC.
• Built an end-to-end Python/PyTorch/TensorFlow package covering data ingestion, training, evaluation, and reporting — handed off to internal scientists as a self-serve tool.

Multi-omics integration for activity-dependent neuroscience.

• Applied supervised ML to investigate activity-dependent changes in the neuronal proteome.
• Designed a multimodal integration and domain-adaptation pipeline bridging proteomic and transcriptomic datasets — foundational experience in cross-modality reasoning later applied to industry.

Real-time perception for an image-guided biopsy robot.

• Developed real-time needle-tip localization and tracking from ultrasound imaging, enabling autonomous robotic control for image-guided biopsy.
• Built a C++/CUDA software interface coupling real-time image analysis with robot control — first exposure to shipping latency-critical scientific software.