• Orchestrated the successful launch of Ionpath’s new repeatable services offering called MIBIplus. This involved overseeing the commercial launch projects across reagents, software development, bioinformatics, and marketing initiatives.• Directed the development of MIBIplus Manager, a user-friendly online portal for project ordering, management, and tracking. Collaborated closely with third-party developers to ensure integration with Ionpath’s cloud-based visualization software.• Managed cross-functional teams, including reagent R&D and software engineering driving improvement of Ionpath’s product offering for both MIBIplus and MIBIscope customers.• Led software release management process, including creating release notes, updating user guides, and coordinating regression and user acceptance testing to ensure smooth software deployments.

• Headed a team of field application scientists, which served as primary point of contact for fifteen existing MIBIscope installation sites and future customers. Collaborated closely with field service and engineering teams to resolve customer hardware and software issues.• Led Reagents R&D team, enhancing the reliability and offerings of the MIBI consumables line.• Provided crucial customer insights during the development of major software and algorithm features for MIBiscope customers that increased system usage and throughput by a factor of 2.

• Reported customer issues and pain points to product, engineering, and software teams, escalating as necessary when recurring issues prevent customers from making reliable measurements. Provided user perspective and stories to R&D teams as features and updates are scoped and implemented.• Designed, implemented, and enforced workflows for support case resolution, escalation, and documentation using Zendesk, Jira, and internal communication tools.• Represented end users in software release cycle planning by advocating for new features, prioritizing existing bugs and issues, performing user acceptance testing of software release candidates, and writing software release notes.• Troubleshooted and resolved customer issues across the entire MIBIscope platform, including assay development, sample preparation, instrumentation, and data analysis.

• Designed and developed desktop application to post-process instrument data, integrating it with cloud-based data management software using a REST API. Productionized the isobaric correction algorithm resulting in a 10x speedup of execution.• Utilized the multiplexed nature of MIBI data to train object detection models using TensorFlow for single cell segmentation with multi-channel input data.• Managed undergraduate level summer and winter interns on the bioinformatics team.• Contributed to open-source Python MIBI image data analysis tools used both in-house and by external users.• Developed an algorithm which leverages spatial nature of interfering signal in MIBI images due to isobaric species created during the measurement. Worked with in-house scientists to design experiments for characterizing, understanding, and quantifying the interference. • Supported contract research at Ionpath by performing downstream analysis of MIBI data, including single cell segmentation, phenotyping, cell-cell distance profiling, and cell density measurements for top pharma companies.• Developed Python code base for internal MIBI data users for data manipulation and visualization.

• Developed an algorithm which leverages spatial nature of interfering signal in MIBI images due to isobaric species created during the measurement. Worked with in-house scientists to design experiments for characterizing, understanding, and quantifying the interference. • Supported contract research at Ionpath by performing downstream analysis of MIBI data, including single cell segmentation, phenotyping, cell-cell distance profiling, and cell density measurements for top pharma companies.• Developed Python code base for internal MIBI data users for data manipulation and visualization.

• Designed, trained, and implemented a fully convolutional neural network with TensorFlow for denoising and removing background signal from mass spectrometry images of uranium-bearing particles. The neural network implemented a machine learning approach to create a predictive model of relevant versus background signal to supersede thresholding techniques that are inherently user biased and not widely applicable. • Compiled a set of mass spectrometry images to serve as a training data set for the fully convolutional neural network. Sought and identified existing images which comprise different types of analyses to train a robust and widely applicable neural network. • Segmentation of SIMS chemical images for single particle measurements using traditional and machine learning based approaches. • Furthered the application of image fusion for detection of nuclear materials with dynamic SIMS images of U-bearing particles using X-ray energy-dispersive spectroscopy (EDS) as a source of higher resolution information.