• Developed a simulation-based congestion forecasting model to estimate the optimal number of Superchargers required to balance customer experience and cost.
• Productionized the Supercharging network planning system, generating monthly forecasts for 5k+ sites and daily development recommendations with an annual budget exceeding $500mn.
• Designed and implemented a scalable geospatial telemetry pipeline using Spark and Airflow, processing 5mn+ daily vehicle trips into hex-level data for diverse downstream applications.
• Built interactive RShiny applications that queried Mongo and SQL databases to map and calculate customer density statistics.
• Used unsupervised learning algorithms (DBSCAN) to identify clusters of customers underserved by current sales and service facilities.

• Developed a real-time electric vehicle energy management system to minimize energy consumed while providing desired acceleration using a multi-objective MDP model implemented in Julia.
• Validated the optimization model in a custom simulation environment showing 10% energy consumption reduction over baseline.
• Collaborated with an interdisciplinary team to deploy and demo the intelligent planner and an auxiliary human-machine interface on a test vehicle for the Head of Global R&D and Nissan COO.

• Designed and produced AI Blindspot Cards to help decision makers ethically implement AI.
• Worked with technologists, managers, and policymakers to confront emerging AI ethics and governance challenges.

• Led a team to develop an advertisement bidding system using Bayesian methods to optimize daily bids and budgets for thousands of campaigns, resulting in a ~13% increase in margin.
• Architected and implemented a brand suitability classifier for YouTube videos using NLP techniques to achieve ~0.92 AUC.
• Conducted user research and programmed a chatbot with Python and an API.AI backend to summarize campaign performance.
• Managed and updated Docker containers for seamless engineering-data science iteration.
• Partnered with business and engineering leadership to define and track data science goals.

• Compared classification models (decision tree, random forest, SVM, Naive Bayes) to evaluate prescriber affinity toward in-person sales channels.
• Used regression models to predict clinical trial enrollment for a $117bn biotechnology company.
• Built an Excel-based database reporting tool for a $93bn pharmaceutical client.

linkedin:
linkedin.com/in/wenjeff
github:
github.com/jeffwen
website:
jeffwen.com

proficient:
Python, R, SQL, Spark, Git
learning:
Processing, Julia, D3.js,
JavaScript, Rust, MongoDB

fluent:
english
cantonese
limited:
mandarin