Gweneth Ge

Gweneth Ge is a master student in the Robotic System Development Program at Carnegie Mellon University, committed to building next-generation automation labs to accelerate drug development. Before joining CMU Robotics Institute, Gweneth had worked at a TCR T-cell therapeutic startup, where she automated high-throughput TCR cloning, screening and validation experiments.

Gweneth's work focuses on protocol generation and optimization algorithms with minimal user input, and improving automated liquid handling systems to be more autonomous, affordable, and robust to achieve end-to-end DNA oragami assembly.

Education

  • M.S. in Robotic System Development, Carnegie Mellon University (Expected May 2026)
  • M.S. in Automated Science in Biological Experiments, Carnegie Mellon University (2023)
  • B.S. in Computer Science, University of California, Irveine (2021)

Research Experience

  • Bioprinted Lung Organoid Perfusion (2023-2025 Research Project) Developing an automated long-term perfusion system for 3D-printed lung organoids, including lumen detection using computer vision and multi-device coordination across the system.

  • DARPA Triage Challenge UAV Platform (2023-2025 Capstone Project) Designed and built a fully NDAA-compliant drone system for search-and-triage in disaster scenarios. Developed UI for rapid drone control, including auto land, take off, arm/disarm, Estop, gimbal control, auto-lock patients for triage and detection.

  • Liquid Handler Protocol Generation
    Developed a domain-specific language and intelligent search algorithms (Beam Search, Genetic Algorithm, A*) to automatically generate robotic protocols from high-level reagent goals.

  • Closed-loop PCR Optimization Built a closed-loop system with active learning to optimize PCR reactions in simulation, significantly improving prediction accuracy for product yield.

  • Cell Painting for Drug Profiling Automated drug treatments and conducting phenotypic profilings of HeLa cells, extracting image features with CellProfiler to analyze morphological affection by different drugs with various doses.

  • Hypsodonty Open Chromatin Analysis Analyzed over 16,000 genes across species to identify regulatory elements linked to hypsodonty, revealing Medium Spiny Neuron D1 as the most relevant cell type.

Publications and Workshops

  • Reinforcement Learning for Robotic Liquid Handler Planning, WABI 2023
    Mohsen Ferdosi, Gweneth Ge, Carl Kingsford
  • ActionRAG: Generalizing Instruction-to-Code via Action Graph Retrieval for Robotic Manipulation, RSS 2025 Workshop, Yufan Liu, Yi Wu, Gweneth Ge, Haoliang Cheng, Rui Liu

Awards

  • James R. Swartz Entrepreneurial Fellowship (2024–2026)
  • Dean’s Fellowship, Carnegie Mellon University (2021–2023)

Contact Information

Please contact Gweneth by email: yuejung@andrew.cmu.edu