About Me

I am interested in the design and governance of trustworthy artificial intelligence systems. My aim is to help people and organizations actively center ethical considerations of AI.

I am currently a Responsible AI Senior Program Manager at Workday as well as a part-time Ph.D. student at Drexel University studying Information Science.

Previously, I worked as a data scientist at Pfizer where I pioneered their enterprise-wide Responsible AI strategy and toolkit, and as a decision analytics associate at ZS Associates. I obtained my bachelor's degree in Information Science and Statistics from Cornell University.

Home: Bio

Education

Philadelphia, PA

September 2022 - Present

Drexel University

Ph.D. Student Information Science

Research: Graduate Research Assistant in Dr. Jina Huh-Yoo's H2 Design Lab.

Courses: Foundations of Human-Centered Computing, Foundations in Data and Information, Foundations of Information Science, Human Computer Interaction

Ithaca, NY

Aug 2012 - Dec 2015

Cornell University

B.A. Information Science, Statistical Science

GPA: 3.5

Dean's List: Fall'12, Fall'14

Teaching Assistant

Data Mining & Machine Learning
Introduction to Programming & Web Design

Activities & Societies

Alpha Phi Omega - Brother
Hindu Student Council - Vice President, Webmaster
Society of Women Engineers - Corporate Relations Liaison
Cornell University College of Arts & Science Ambassadors

Home: Education

Research Interests & Projects

I am a graduate research assistant in the Drexel Information Science Department. I am a member of Dr. Jina Huh-Yoo's H2 Design Lab. Our lab is focused on researching the role of technological solutions in health-related social contexts.

Community-based Telehealth for Older Adults

We are working with the Telehealth Intervention Program for Seniors (TIPS) program. TIPS is a community-based telehealth program in the tri-state area (New Jersey, New York, Pennsylvania). The program allows older adults to regularly monitor biometric measurements, such as blood pressure, heart rate, weight, etc, and self-reported outcomes. Abnormal readings trigger alerts that are triaged and managed by remote nursing staff, which help to escalate to participants' providers.

My first research project focused on identifying factors that led to disparities in the time remote nurses take to respond to incoming alerts. Our paper is under review and is available for pre-print.

Home: Projects

Professional Experience

May 2022- Present

Responsible AI Senior Program Manager, Workday

Piloted the update of Workday’s ethical AI principles.
- Conducted a benchmarking analysis of 48 private and public sector AI ethics principles and advances in Workday’s Responsible AI development practices.
- Leveraged text and graph analytics to analyze AI principles text data to form recommendations.
Designed Workday’s Responsible AI Risk Evaluation process to systematically identify the risk level of Workday’s AI feature.
- Researched existing and emerging regulations and standards related to AI Risk management and designed into a simple questionnaire for product managers to easily incorporate into their workflows.
- Worked with cross-functional team members from product, machine learning, and legal teams to incorporate their feedback to improve the risk evaluation process.
Developed a standardized approach to conduct fairness testing on Workday’s highest risk AI features.
- Led a cross-functional workshop with members from Workday’s machine learning, legal, compliance, and product teams to establish a holistic approach to testing that comprises of legal standards for adverse impact testing as well as causal inference methods to identify potential sources of biases.
- Defined the governance to provide structure for fairness testing including testing requirements during the product lifecycle, testing evaluation criteria, and roles and responsibilities to conduct and review testing results.
- Conducted fairness testing on Workday’s highest risk features
Launched Workday’s Responsible AI research and engineering arm to bring AI research and engineering expertise to the team
- Led hiring for Workday’s Responsible AI research scientist
Providing subject matter expertise for several Responsible AI efforts including training and awareness initiatives, maintenance of fact sheets to provide transparency and documentation of our AI features for customers, and company policies related to artificial intelligence.

New York, NY

Nov 2017 - Apr 2022

Data Science Manager, Pfizer

PMO-lead for the development of Pfizer’s first Responsible AI enterprise framework to address the ethical, legal, and regulatory considerations of artificial intelligence
- Leveraging Agile framework to coordinate creation of cross-functional solutions including legal policies, governance risk processes, company-wide protocols controls, ethical AI R&D solutions, and communications & change management
Developed and deployed Pfizer’s first Ethical AI toolkit for Pfizer data scientists focusing on bias detection and model transparency
- Designed Jupyter notebook templates, custom Dataiku plugins for Pfizer’s data science platform, and Tableau dashboard templates for data scientists to easily incorporate into their workflows
Data Science lead for multiple predictive analytics workstreams to understand patient outcomes with real world data, open-source data and internal sales and marketing customer datasets across multiple therapeutic areas
- Lead machine learning algorithm development of Pfizer’s first publicly deployed AI-based digital companion for physicians to estimate the likelihood of a rare heart condition: https://estimattr.com.hk/
- Developed ensemble machine learning model forecast of COVID-19 cases to assist COVID-19 Phase 2,3 trial recruitment in the US.
Engaged in internal and external speaking opportunities to increase awareness for Responsible AI
- TEDxPfizer 2021 speaker - Can we trust artificial intelligence (Proprietary)
- Dataiku Product Days 2021 speaker – Utilizing Dataiku to scale AI Ethics
Established and co-lead the talent pipeline hiring.
- Hired 5 full-time team members for the Pfizer Thessaloniki Center for Digital Innovation Hub established in 2020
- Led hiring for 3 full-time team members in Pfizer NYHQ
- Co-led the summer student worker program and co-op program (2019-present). Hired 20 summer student workers and fall co-op students over three years

Volunteer on outreach programs such as Girls Who Code, BT 4 Girls, and WiTNY to encourage young girls to consider careers in STEM and analytics

Philadelphia, PA
Feb 2016 - Nov 2017

Decision Analytics Associate, ZS Associates

Partnered with several clients across the pharmaceutical and asset management industries to design predictive modeling frameworks. • Used R and SAS to develop marketing mix (or promotional response) models, customer targeting, and customer attrition models
Trained the client teams in R to continue use of delivered analytics solution
Managed team of onshore interns and offshore resources

Home: Experience

Publications & Presentations

2023

Journal of Cardiac Failure

Castaño, A., Heitner, S. B., Masri, A., Huda, A., Calambur, V., Bruno, M., ... & Shah, S. J. (2023). EstimATTR: A Simplified, Machine-Learning-Based Tool to Predict the Risk of Wild-Type Transthyretin Amyloid Cardiomyopathy. Journal of Cardiac Failure.

2023

American Medical Informatics Association Annual Symposium

Kritharidou M., Chrysogonidis G., Ventouris T., Tsarapastsanis V., Aristeridou D., Karatzia A., Calambur V., Huda A., & Hsueh S. (2023). Ethicara for Responsible AI in Healthcare: A System for Bias Detection and AI Risk Management. In AMIA.

2022

American Medical Informatics Association Annual Symposium

Aristeridou D., Calambur V., Mazzetta B., Ateya M., Haque S., Colvecchia C. (2022, November 5-9) Validation of a Two-year Risk Prediction Model for Undiagnosed Atrial Fibrillation Using National Electronic Health Record Data. Poster presented: American Medical Informatics Association (AMIA) Annual Symposium, Washington, DC.

2020

International Society of Amyloidosis

Huda A., Heitner S., Calambur V., Bruno M., Schumacher J., Emir B., Isherwood C., Castaño A. 2020. A Machine Learning Framework for Predicting Risk Of Wild-Type Transthyretin Amyloid Cardiomyopathy. International Society of Amyloidosis (ISA)

2020

American Heart Association Scientific Sessions

Heitner S., Masri A., Elman MR., Emir B., Nolen KD, Schumacher J., Calambur V., Huda A., Bruno M., Castaño A. 2020. A Simplified Machine Learning Algorithm for Identification of Patients at Risk for Wild-Type Transthyretin Amyloid Cardiomyopathy. American Heart Association (AHA) Scientific Sessions 2020

2021

Dataiku Product Days Speaker

Utilizing Dataiku to scale AI Ethics explores that hot question — “Can we trust artificial intelligence in healthcare?” — and to tell us about the ways that Pfizer is navigating the AI ethics conversation with Dataiku. Algorithm fairness and transparency are primary priorities for AI ethics at Pfizer, and they have built out an AI Ethics Toolkit which utilizes Dataiku for bias detection, bias mitigation, and model transparency.

2022

60 Leaders on AI

From Algorithms to Ethics. A holistic view of AI. Contributed to Chapter 11 - What are the ethical concerns regarding the massive power and the general adoption of AI?https://www.60leaders.com/leader/veena-calambur?s=book

Home: Experience

Awards & Recognition

2022

NIST AI Risk Management Framework Workshop Panelist

Panel speaker during NIST AI Risk Management Framework Workshop #3. Provided feedback on the Management section of the NIST AI Risk Management Framework draft in advance of the publication of the standard.

2020

Reuters Pharma Awards: Most Promising Agile Transformation

This award recognizes a more innovative and entrepreneurial culture, either through creation of a new incubator/in-house startup initiative, or embedding of agile and iterative processes across major parts of the company. We are particularly keen to see ‘learning organizations’ with new metrics, or major improvements in data transparency and sharing, ultimately leading to faster, tailored, quantified benefits for HCPs, patients or other stakeholders.

2020