Jonathan Potter
Automation Engineer | Pioneering the Future of Scientific Discovery through Robotics and AI
Get in TouchAbout Me
I'm an Automation Engineer and graduate student at Carnegie Mellon University, studying Automated Science. My passion lies in revolutionizing scientific research through cutting-edge technology, particularly in the realms of drug discovery, genomic research, and longevity studies.
With a background in Biological Systems Engineering and expertise in robotic liquid handling systems, active learning frameworks, and custom labware design, I'm dedicated to empowering scientists by automating time-consuming processes. My goal is to introduce the first truly autonomous robotic scientists, capable of conducting research around the clock and in parallel, pushing the boundaries of what's possible in scientific discovery.
Skills
- Robotic Liquid Handling (OpenTrons, ThermoFisher Momentum, CyBio Felix)
- Machine Learning & Active Learning
- Python, R, MATLAB, Go, Arduino C++
- 3D Printing & CAD for Custom Labware
- Lab Automation
- Bioinformatics
- Data Analysis
- Molecular Biology
Past Work
Automation Engineer
Magnify Bioscience, Pittsburgh, PA
September 2024 - Present
- Automated a six-hour chemical process using Opentrons V2 robotic liquid handlers and heater-shaker modules, reclaiming hours of walkaway time
- Designed and implemented custom 3D printed hardware into the OT-2 protocol to improve efficiency and quality of results
- Created a user-friendly GUI enabling lab members without programming experience to execute the protocol
Bioinformatics Intern
Predictive Oncology, Pittsburgh, PA
May 2024 - August 2024
- Architected and developed a software package that combines patient and treatment information stored in a LIMS database to strategize plate layouts and construct scripts for liquid handlers
- Developed a fully-portable automated script generation pipeline enabling a smooth transition between old and modern liquid handlers
- Performed extensive performance analysis to present the capabilities and limitations of various hardware choices
Bioinformatics Intern
Collins Lab at UC Davis, Davis, CA
April 2020 - August 2023
- Leveraged CellPose, Numpy, and TensorFlow to build innovative cell-specific machine-learning models to perform quantitative microscopy on imaging data
- Restructured and developed an automated pipetting robot, employing custom 3D printed parts and Python
- Shared findings in weekly lab meetings and worked with fellow life scientists and biostatisticians to determine future directions for experimentation
Projects
Corteχ
Created a software package to streamline patient selection for clinical trials by processing complex clinical trial criteria and matching patients based on their data. This application leverages natural language processing and a rules engine to assess patient suitability, providing scores and exclusion reasons based on specified inclusion and exclusion criteria.
Corteχ was designed in 24 hours for the Nucleate Pittburgh 2024 Biohackathon, and is available on GitHub here:
CorteXCodex
A software package optimizing plate and liquid handler deck layouts for maximum time efficiency in robotic operations. Codex aims to revolutionize how labs plan and execute automated experiments.
Automated Snake Antivenom Gene Classification
Led a project automating the experimental classification and performance evaluation of snake antivenom genes, streamlining the research process in this critical field.
Cancer Subtyping with Machine Learning
Collaborated on a project using machine learning techniques to subtype cancer based on single-cell RNA sequencing data, contributing to more precise diagnostics and treatment strategies.
Vision for Automated Science
My vision for the future of scientific research is one where automation and human ingenuity work hand in hand. By leveraging cutting-edge technology, we can free scientists from time-consuming, repetitive tasks, allowing them to focus on what they do best: asking the right questions, designing innovative experiments, and making groundbreaking discoveries.
As I often say, "Let robots do the menial everyday tasks, so that scientists have more hours in the day to actually do science, not chores." This approach doesn't aim to replace scientists but to empower them. By automating routine procedures, we can increase experimental throughput, reduce human error, and unlock new possibilities in research design and data analysis.
I'm excited to be part of this transformation, developing tools and methodologies that enhance the capabilities of research teams across various fields, from drug discovery to genomic research and beyond. Together, we can accelerate the pace of scientific progress and tackle some of the most pressing challenges facing humanity.
