Washington, District of Columbia, United States
I'm fascinated by data. It confirms, it denies, it predicts, it influences, it surprises. It requires intent and curiosity to work with it meaningfully. It takes empathy and awareness to distill it down to help others understand its story. Every day, I work with data to develop, improve, and deliver successful outcomes and data-driven products in challenging domains.
• Drive supply chain operational excellence for Google Data Center Equipment Supply Chain (Google Cloud). • Lead high-impact, complex, multi-disciplinary data management & engineering projects. • Plan requirements with internal customers and usher projects through the entire project lifecycle. • Manage project schedules, identify risks, and clearly communicate with project stakeholders. • Analyze technical trade-offs and recommend product development solutions to executive leadership.
Overall Experience: • Consult for a high-profile tech company (one of the Big 5), driving data strategy & building BI tools for supply chain & vendor management programs to improve global data center construction outcomes. Management Experience: • Mentor & coach cross-functional team members (technical staff, non-technical staff, & interns). • Facilitate regular team meetings to offer guidance & support, with the aim to create team autonomy. • Conduct weekly 1:1's with staff with the intent to coach, give feedback, & share ideas in a safe forum. • Structure project plans, track individual task progress, and track overall team progress using Asana. Client Experience: • Identify & recommend high-value project opportunities that align with client OKRs. • Cultivate client relationships with regular check-ins to align on expectations & ensure service quality. • Develop long-term strategies & present roadmaps to senior leadership to add value to the business. Technical Experience: • Manage various data pipelines and BI workflows, including data collection from disparate sources, data hygiene, data warehousing, ad-hoc analytics, dashboard development, and process automation. • Write scripts in GoogleSQL (BigQuery) & Python (Colab notebooks). • Conduct SQL & Python code reviews. • Provide spreadsheet expertise to others across the organization as needed (writing formulas, etc.).
• Managed data science and data analytics workstreams to improve business outcomes through taxpayer notice prototype testing at the IRS. • Managed efforts of a team of highly technical data analysts and strategic business consultants. • Ran experiments to test the effect of behavioral insights treatment on notices, optimizing for compliance actions and self-service resolution. • Analyzed characteristics and journey patterns on millions of taxpayers to explain behavior and identify improvement areas. • Trained propensity models to predict the probability of payment if sent a notice, selecting the most “persuadable” taxpayers to get a notice to influence compliance behavior. • Wrote data pipeline scripts in Python and SQL to automate queries, summaries, and visualizations for consistent and reproducible analyses. • Designed and presented critical data insights via slide content to IRS executives and stakeholders in a fast-paced environment.
• Used data science, data analytics, randomized controlled trials, and behavioral insights research to predict, understand, monitor, and explain the effect of taxpayer notice redesigns, optimized for IRS business objectives. • Wrote functionalized scripts and pipelines in Python and SQL to automate querying, cleaning, transformation, analysis, and visualization of data from relational databases containing billions of records – for consistent and reproducible analytics. • Generated interactive HTML plots using Plotly to populate a data dashboard for IRS executives to monitor notice performance metrics. • Business management consulting to IRS to develop strategic plans to improve the process of generating notice correspondence – as well as a plan for the prioritization and phasing of redesigning hundreds of notice products. • Task management for several projects to plan and monitor task progress against backlogs of major tasks and milestones. • Delivered insights to IRS executives and stakeholders regularly through rapidly developed presentations, often on very tight timelines. • Trained models predict transaction fraud. Coded a functionalized end-to-end pipeline to read in, clean, join, and transform millions of observations to pass through the selected anomaly detection model and generate predictions.
• East coast liaison for firmwide data strategy efforts to streamline data collection processes and incorporate tools for data-driven insights in daily architectural processes. Evangelized analytics, data science, and machine learning endeavors through presentations, initiatives, and case studies. Interfaced with project managers to better understand how data can help their decision-making, team productivity, and staffing. Worked to streamline data collection processes. • Developed a client product demo in Streamlit for a computer vision classification tool using a convolutional neural network with transfer learning using Keras with Tensorflow backend to detect proper interior finishes per company branding guidelines. Developed an NLP algorithm using NLTK in Python to extract general user sentiment from open-ended survey questions to identify recurring patterns or trends. (Such as acoustics, temperature, light, etc...) • Generated dashboards in Power BI that identified trends in software usage firmwide, focusing in particular on Autodesk Revit usage and model sync time per user. These metrics identified more heavily used versions of Revit software to control cost, understand usage patterns, and proactively find users having syncing problems. • Queried and joined tables from sizable SQL databases to aggregate measures across various dimensions and groups – such as total time spent using Revit software per office, per project team, per day – to identify if a project team or office is more overworked than others. • Designed, deployed, and analyzed occupant survey data from clients & building users to better understand what people like and don't like in the spaces they use, to inform future design decisions. • Managed project teams with internal teammates and external consultants. Built Dynamo scripts to automate various Revit tasks. Manipulated Revit data using Excel & Ideate BIM Link - and exported to Ecodomus to format data for COBie deliverables.
• Freelance writer regularly publishing content related to data science and machine learning to the Towards Data Science publication on Medium. • Towards Data Science (TDS) provides a platform for thousands of people to exchange ideas and to expand our understanding of data science. The audience is mixed, consisting of readers entirely new to the subject and expert professionals who want to share their inventions and discoveries. • Writing technical, conceptual, and product-related blogs, with a goal to discuss and shed light on important concepts and developments in data science – and share new ideas that I have to use the power of data science to solve real-world problems.