Cambridge, Massachusetts, United States
I work at the edge between what medicine thinks is possible with data and what actually is. Most organizations sit on data that could tell them which patients will respond to which treatments, which trials are enrolling the wrong population, which biomarkers are being ignored because nobody connected the right datasets. The bottleneck isn't AI capability. It's data that's fit for purpose — structured to capture what biology actually needs to say, accessible at the speed decisions need to be made, and trusted by the people who have to act on it. That's the problem I solve. As one of the first data hires at Foundation Medicine, I rearchitected the pharmacogenomic data model from a system that couldn't scale to one that multiplied scientist throughput 10x — and built adoption by making the better path the easier path. That data foundation survived Roche acquisition due diligence. At BMS I lead a data engineering team building the data layer that makes foundational model development possible — unifying tens of millions of patient biosamples, enabling massive parallel data access, and shifting R&D culture toward data-first decision making. I think in connections and links. I see where knowledge graph architecture, FAIR data principles, and semantic engineering can unlock decisions that organizations didn't know they could make. And I've learned that the best data infrastructure is the kind people actually use. W3C contributor · 40+ publications · MIT Grand Hack winner · PhD Bioinformatics
As a Manager at ZS, I lead teams that work on projects with clients to build ontologies and knowledge graphs. Our team utilizes ETL tools, both using industry standards and customizing to client needs, to enable the creation of powerful knowledge graphs. I assisted clients in understanding the impact of federating queries and help them create knowledge graphs from scientific text using bioBERT and GPT APIs, as well as maximize the tools they already paid for such as Domino, Collibra, Benchling and others. These graphs offered powerful backends for several applications in drug discovery, target discovery, and identifying new genes for specific diseases. Additionally, I supported finding combination therapies for CAR-T via literature extraction and assisted in antibody engineering by connecting assay data with databases of antibody components with ontologies and information architecture. I have expanded my work to include medicinal chemistry and product development with taxonomies and ontologies.
As a seasoned team leader, I have a track record of driving technical improvements and fostering a positive work culture. One notable project was leading a team to improve the backend of ClinicalKey, a flagship medical documentation search tool. I encouraged the team to develop in a test-driven fashion and learn devops to ensure the code was production-ready. My efforts to promote architecture approval and thorough software security review resulted in a robust and secure solution. I learned about the importance of creating a listening environment to ensure that issues are raised early and addressed before becoming larger. This was reflected in my leadership style and the positive results that my team achieved. I also contributed to the field of Translational Medicine by publishing a paper on the subject. In response to the COVID-19 pandemic, I quickly assembled a team to build a secure web portal for COVID researchers to access scientific data for free. The portal was ready in just two weeks, and has contributed to the creation of a vaccine. My leadership during this project earned me recognition in the form of a "silver coin."
In this role, I have established a proven track record of building strong partnerships between IT and business experts. Through my expertise in data science, I have been able to empower data scientists by providing them with access to training data through partnerships with crowdsourcing tools like Figure-Eight and Mechanical Turk, as well as working closely with internal SME teams. My passion for data science has also helped the firm establish a reputation for thought leadership in the field. During this time, I have been a regular speaker at conferences on data science and natural language understanding (NLU), delivering four invited keynote addresses. In addition, I was instrumental in the company's decision to acquire 3D4Medical, having written an internal white paper that compared various VR and AR technologies and informed the decision-making process. Through hands-on delivery and teaching others (including executives) I have developed a deep understanding of advanced technologies such as deep learning, machine vision, knowledge graphs, and parallel computing, and have used these tools to help scientists make breakthrough discoveries. Additionally, I have applied both supervised and unsupervised machine learning techniques to extract knowledge nuggets from millions of Elsevier papers. Furthermore, my experience in parallelizing big data pre-processing techniques through PySpark has allowed me to tackle even the most complex data science challenges.
In my role as a Coordinator for the Foundation Medicine Knowledge Base (FoundationCORE), I was responsible for ensuring the on-time delivery of this critical project. In addition to technical leadership of the products I had developed in the previous years, my role involved evaluating text mining and natural language processing software provided by third-party entities and working with them to enhance Foundation Medicine's biomedical knowledge extraction capabilities. I was also a pioneer in driving the data sciences effort within the company and led several initiatives aimed at improving the company's capabilities in this space. My work on mapping the various data sources, teams, dependencies and software tools into a master data management flow diagram helped lay the foundation for the company's data governance practices. I also developed APIs that supported products aimed at enhancing the utility of the company's molecular information knowledge base, such as PatientMatch and GeneKit. In addition to my technical skills, I developed strong project management abilities and have a proven track record of successfully coordinating the requirements gathering, design, and project management of complex software development projects. My experience leading a cross-functional team and collaborating with internal stakeholders to drive projects forward is a testament to my ability to lead initiatives and drive positive outcomes.
As a data sciences expert in precision medicine, I have made significant contributions to the field through my research in big data, smart data, and linked data solutions. My expertise in these areas allowed me to lead the development of precision medicine decision support software for cancer, resulting in the filing of a patent on FoundationCORE. My passion for linking data and semantic web technologies was a key factor in driving company-wide adoption of these innovative approaches. I served as a trusted liaison between technology and the biomedical, pathology, and medical affairs groups, ensuring seamless integration of these cutting-edge technologies into the company's operations. In addition, I identified and implemented a bioinformatics algorithm for the assignment of genomic alterations to clinically relevant alteration groups, further enhancing the company's ability to provide precise and effective care to patients.