Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

AI Engineer with experience across computer vision, machine learning, 3D reconstruction, and generative AI, applied to medical imaging and agricultural use cases. Strong end-to-end background from dataset preparation and model development through cloud training, deployment, and monitoring. Known for turning research ideas into practical systems, building reliable data/ETL pipelines for imaging and 3D data, and designing agentic LLM workflows and hybrid retrieval architectures for high-precision outcomes.…AI Engineer with experience across computer vision, machine learning, 3D reconstruction, and generative AI, applied to medical imaging and agricultural use cases. Strong end-to-end background from dataset preparation and model development through cloud training, deployment, and monitoring. Known for turning research ideas into practical systems, building reliable data/ETL pipelines for imaging and 3D data, and designing agentic LLM workflows and hybrid retrieval architectures for high-precision outcomes.

Yang Yang

AI Engineer, Data Scientist, AI Developer, +1





AI Engineer with experience across computer vision, machine learning, 3D reconstruction, and generative AI, applied to medical imaging and agricultural use cases. Strong end-to-end background from dataset preparation and model development through cloud training, deployment, and monitoring. Known for turning research ideas into practical systems, building reliable data/ETL pipelines for imaging and 3D data, and designing agentic LLM workflows and hybrid retrieval architectures for high-precision outcomes.…AI Engineer with experience across computer vision, machine learning, 3D reconstruction, and generative AI, applied to medical imaging and agricultural use cases. Strong end-to-end background from dataset preparation and model development through cloud training, deployment, and monitoring. Known for turning research ideas into practical systems, building reliable data/ETL pipelines for imaging and 3D data, and designing agentic LLM workflows and hybrid retrieval architectures for high-precision outcomes.

Available to hire

AI Engineer with experience across computer vision, machine learning, 3D reconstruction, and generative AI, applied to medical imaging and agricultural use cases. Strong end-to-end background from dataset preparation and model development through cloud training, deployment, and monitoring.

Known for turning research ideas into practical systems, building reliable data/ETL pipelines for imaging and 3D data, and designing agentic LLM workflows and hybrid retrieval architectures for high-precision outcomes.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Beginner

Beginner

Work Experience

AI Engineer at Formus Labs

January 1, 2026 - Present

Developed a new 3D reconstruction approach for hip bones from biplanar X-rays, enabling generation of the internal cortical surface. Achieved ~2mm accuracy on synthetic X-ray data and progressed on real X-ray reconstruction. Partnered with the CEO to define the X-ray data acquisition protocol. Trained models on AWS A100 instances, optimized training workflows to reduce time from 7 days to 3 days, and deployed the model on Google Cloud. Processed medical imaging data (DICOM/NIfTI) for CT and X-rays including cleaning, labeling, orientation handling, synthetic data generation, and dataset preparation. Reviewed literature to guide implementation decisions.

Research Associate at University of Canterbury

March 1, 2024 - March 1, 2025

Built a 3D fruit counting and instance segmentation pipeline using 3D Gaussian Splatting and contrastive learning; improved separation of adjacent apples by redesigning the loss function. Benchmarked and integrated multiple 2D detectors (Detectron2, YOLOv8, Detrex) to guide 3D segmentation. Implemented distributed hyperparameter tuning across 300+ machines using Optuna and Redis Queue. Created monitoring dashboards with Grafana for training progress; built a custom 3D visualizer (Qt, PyRender) for analyzing segmentation results in 3D. Explored optimizer modifications and reviewed relevant 3D instance segmentation literature to improve model performance.

Intern at Fruitminder

October 1, 2023 - March 1, 2024

Engineered a cross-modal 2D-guided 3D segmentation framework by projecting 2D semantic masks from foundational models (GroundingDINO, SAM) onto 3D Gaussian fields to isolate target objects (fruit trees) from complex backgrounds. Implemented supporting tooling and workflows using 3D Gaussian Splatting, Open3D, and CUDA-enabled components for processing and prototyping.