Nicolás Aramayo Benvenutto

The first fulfillment, crossdocking and last mile delivery sharing economy for e-commerce

• Our research work "A Multi-Armed Bandit Approach for House Ads Decisions" was presented at the Khipu Latin American Meeting In Artificial Intelligence in November, 2019 • Led the creation of the company's first AI-based SaaS product for sales executives in the financial industry • Developed a prototype to automate the layout design of banner images for retailers • Designed and taught a class for structuring and developing machine learning projects for Project Managers

• Display advertising optimization via reinforcement learning techniques in A/B test experiments for the home page of one of the largest retailers in the region • Predictive reactivation of churned customers by calls made by sales executives measured in an A/B test experiment ran simultaneously in 3 countries • Design of a reinforcement learning algorithm to sequentially allocate digital marketing budget to optimize conversion of leads • Automatic categorization engine for classifying new products in large category trees for retailers • Chatbot created from scratch for financial institution that dealt mostly with colloquial language and had the ability to store memory during conversations to respond the following questions

• Assisted in the development of the company's core recommendations system in aligning it with its value proposition and its posterior implementation, which was an integer programming model integrated with a probability of purchase model to recommend an optimal mix of products for each salesman in each visit to corner stores • Helped build the data science team experiment design methodology and capabilities

• Research on decision making of different agents in educational systems, structural modeling, and causal impact/counterfactual analyses of public policies. • Our resulting publication "Better School versus Right School? Heterogeneous Preferences in School Matching with Price Information" was presented at the Journal of Marketing Research Special Issue: Education and Marketing conference, 2018.

• Implementation of the matching mechanism for the new school admissions system for public and subsidized private schools in Chile for the years 2017 (participating 5 regions of the country and over 77,000 students) and 2018 (15 regions and almost 275,000 students). • Our resulting publication "School Choice in Chile" was published in the journal Operations Research, and was nominated for the EURO Excellence in Practice Award in 2019 and was the runner-up for the IFORS Prize for OR in Development in 2021.

• Probabilistic Machine Learning - Undergraduate and graduate (autumn 2018) • Logistics and Production - Graduate (autumn 2018) • Marketing II - Undergraduate (autumn 2017, spring 2017, autumn 2018) • Marketing I - Undergraduate (spring 2017) • Supply Chain Design - Undergraduate (autumn 2017)