Cleveland, Ohio, United States
• Developed deep learning pipelines for tissue segmentation, pathology classification, and virtual staining — improving data throughput and enabling faster experimental iteration.
• Engineered multimodal AI systems that combine vision transformers (ViTs), GANs, and diffusion-based architectures for high-resolution bioimage interpretation.
• Built scalable inference infrastructure leveraging FastAPI, PyTorch, and Azure Cloud to serve large 3D/4D imaging models with optimized latency and memory efficiency.
• Integrated large language model (LLM)-based tooling to assist in annotation management, metadata generation, and image-to-text interpretation — bridging data science and domain expertise.
• Designed end-to-end visualization and analytics pipelines with Next.js, WebGL/vtk.js, and Python microservices, enabling interactive exploration of multi-GB cryo-imaging datasets.
• Led system design for streaming and rendering pipelines that reduced dataset load times from minutes to seconds, powering real-time scientific review and decision-making.
