Zhangbo Cheng

Applied automation, analytics, and predictive modeling to build data-driven systems for logistics execution and capacity planning. Owned end-to-end operational data pipelines across 14 lanes and 30+ dock gates, tracking ~200 pallets/day on average (300+ at peak) and transforming raw operational events into structured datasets for downstream decision-making. Built and deployed multiple production-grade automation systems that replaced manual, error-prone workflows and laid the foundation for ML-driven planning: Designed a fully automated BOL generation pipeline, reducing daily workflows from 2–3 operators × 5–15 minutes to a single <3-minute automated run, supporting ~20 BOLs/day and ~40 pre-creations/day with near-zero human intervention. Developed web-based data processing tools for transfer-order aggregation and dispatch reporting, reducing processing time by 65%+ and eliminating recurring data inconsistencies and rework. Currently extending this work into machine learning–based forecasting, using historical shipment and dispatch data to model time-series demand for daily order volume and required truck capacity. This includes feature engineering (time-of-day, lane-level patterns, peak effects), baseline statistical models, and ML regressors to support proactive, data-driven dispatch planning. Focused on building deployable, maintainable ML-ready systems, combining Python-based data pipelines, automation, and predictive modeling to bridge real-world operations with scalable machine learning solutions.

Graduate Level Course: Statistical Analysis (BU.510.601) Non-Linear Econometrics for Finance (BU.232.630) Big Data Machine Learning (BU.520.710) Responsible AI (BU.330.735) Empirical Finance (BU.232.640) Machine Learning for Finance (BU.232.775) MBA Course: AI Essentials for Business (BU.520.710) Business Analytics (BU.520.601) Data Science: Artificial Intelligence (BU.920.624) AI Essentials for Business (BU.520.710) Machine Learning for Management(BU.142.775) Executive Education Program: Leveraging AI For Business Success

• Engaged with prospective students through Unibuddy, providing insights and guidance on the Carey experience. • Assisted in organizing and participating in recruitment events, fostering a welcoming environment for new students. • Transitioned to an Alumni Ambassador role, engaging with alumni and supporting alumni events and initiatives.

Technical Lead for early-stage startup CARE (Clinical Trial AI-Assisted Patient Recruitment and Engagement), leading the development of AI-powered patient recruitment solutions. Designed and implemented the system architecture integrating clinical trial datasets with LLM-based reasoning. Built pipeline including ChatGPT-based knowledge elicitation, medical expert verification, fine-tuning of DeepSeek R1, and knowledge distillation for scalable deployment. Collaborated with clinical and business leads to ensure compliance, scalability, and real-world applicability of the platform. Focused on revolutionizing clinical trial recruitment by improving efficiency, reducing delays, and enhancing patient engagement.

Conducted comprehensive literature reviews on vehicle speed estimation using video surveillance, focusing on techniques like optical flow and feature tracking. Implemented and fine-tuned computer vision models, such as YOLO for object detection and Lucas-Kanade optical flow for tracking, to accurately estimate vehicle speeds from surveillance footage. Evaluated and integrated various feature detection methods, including Shi-Tomasi corner detection and FAST, to improve the robustness and accuracy of vehicle tracking. Developed and optimized code to enhance the detection and tracking performance, ensuring reliable and continuous tracking of vehicles in real-time scenarios. Collaborated with senior researchers and contributed to the preparation of research findings for publication and presentations.

Educational Content Creation: Developed in-depth tutorials and blog posts on a variety of Machine Learning topics, including Natural Language Processing, Computer Vision, and Reinforcement Learning. Practical Implementation: Built and tested proof-of-concept ML models (both traditional ML and Deep Learning approaches) to illustrate best practices for real-world applications. Project Leadership: Collaborated on content strategy and project timelines, ensuring the on-time delivery of high-quality educational resources. Technical Writing & Communication: Simplified complex ML concepts into digestible formats—textual explanations, code examples, and interactive notebooks—tailored for both beginner and advanced audiences. Peer Collaboration: Worked closely with fellow interns and senior researchers to refine drafts, troubleshoot technical challenges, and maintain coherent and consistent instructional design. Through these responsibilities, I successfully helped expand AIML’s knowledge base, reaching a broader audience and enabling learners to transition smoothly from theory to hands-on practice.

Data Analysis Research Assistant @ CCVL Research Group, Computer Science Department

Engineered faculty and center websites, enhancing user experience and design, inspired by top academic platforms. Led development of human-AI interaction simulations and GPT-like chatbot systems, optimizing AI-driven communications. Utilized TypeScript, GitHub, Azure, and OpenAI API, driving forward digital health and AI research initiatives.

Minor: Mathematics