Audible is a subsidiary of Amazon that sells and produces spoken audio entertainment, information, and educational programming. Audible sells digital audiobooks, radio, and TV programs, and audio versions of magazines and newspapers. Audible has also become the world's largest producer of downloadable audiobooks. Audible is the United States' largest audio book producer and retailer.

During the summer of 2018, I worked at Audible as an intern on the Romance team in Newark, NJ. The Romance team works on features for the Audible Romance package, which is a type of subscription to which Audible customers may sign up for a fixed monthly price. Once subscribed, Audible Romance customers have access to all Audible Romance books and can check out as many as they would like at no additional cost, for up to 10 books at a time. As an intern on this team, I became familiar with a wide variety of Amazon's internal tools and developed a new feature from backend to frontend. The feature that I worked on for my internship project was an indicator that would show Audible Romance customers the date on which they returned a book; this was something that customers had been requesting for a while; since subscribing to Audible Romance allows them to check out and return as many books as they like, the most avid readers often forget which books they have already read. Customers suggested that being able to see when they most recently returned a book would help them remember whether they had checked out the book before and jog their memory regarding how much of the book they had read.

My project touched upon databases, user experience design, and numerous Amazon internal services. I worked with several other teams and senior software developers, gaining valuable communication skills and becoming familiar with professional coding practices. With the help of my mentor and my team, I surmounted and learned strategies for dealing with technical challenges, such as working with a large and live codebase (in which packages are constantly altered or even deprecated). One neat programming practice that I learned about was automatic code generation; through my internship, I learned how to generate and use Java Service Pages, which in turn generated full webpages, as well as how to use automatic programming to generate service packages. Towards the end of my internship, as I was finishing up my project, I deployed my feature to a test environment and learned how to measure and mitigate the impact that my changes would have on the efficiency of the official Audible website.

Pivotal Software, Inc. is a software and services company which has divisions that perform consulting services (Pivotal Labs), work on cloud computing (Pivotal Cloud Foundry), and develop other big data products. One of these products is Pivotal GemFire, which is an in-memory distributed data grid for high-scale custom applications. GemFire is powered by Apache Geode, which is an open-source data management platform that provides real-time, consistent access to data-intensive applications throughout widely distributed cloud architectures.

During the summer after my sophomore year of college, I worked at Pivotal as an intern on the GemFire Management and Monitoring team. During this time, I worked on Pivotal GemFire and Apache Geode. I had the opportunity to pair-program with my teammates to squash bugs and develop new features and I learned the ropes of how a software company like Pivotal is operated. I participated in a workshop organized and run by Elisabeth Hendrickson, Pivotal's Vice President of Data Research and Development, which dove into agile development and how it can improve the productivity of any workplace. For three weeks I also worked with the refactoring guru Michael Feathers, author of Working Effectively with Legacy Code and Brutal Refactoring: More Working Effectively with Legacy Code, to improve the structure of the Apache Geode and Pivotal GemFire codebases.

While I was not assigned a specific project for my internship, I did have many opportunities to test and develop software for Pivotal. The work I did was often challenging and engaging, and I was always surrounded by people who were experts in agile development, concurrent systems testing, and intricacies regarding Pivotal's big data products. It was an honor to be able to contribute to Apache Geode, an open-source project that people all over the world have worked on, as well as Pivotal GemFire, a data management platform that is used by companies like GE and Indian Railways.

Since Apache Geode is open-sourced, its source code is publicly available. If you're curious, you can take a look at it on GitHub here.

At Olin College of Engineering, a fundamental part of our philosophy is that learning occurs through immersion in real-world applications. The final project that Olin seniors work on is a corporate consulting project, called the Senior Capstone Program in Engineering (SCOPE). SCOPE is a unique industry-university collaboration; over the course of a full academic year, seniors work in multi-disciplinary teams to provide innovative solutions to a company’s real-world problems. Sponsors retain full rights to all intellectual property developed by the team.

Toyota Motor North America's Social Innovation division seeks to make mobility a source of inspiration and a way to improve the quality of life for everyone. As part of this goal, Toyota partnered with my SCOPE team of five senior engineering students to add sensing and control functionality to power wheelchairs. We are collaboratively designing a solution with power wheelchair users to increase their independence in indoor environments.

On our team, I act as a co-product owner with one other team member, and together, we lead and direct the productivity of our team. We have completed significant user research in the first semester of SCOPE and our final semester will involve rapid prototyping and continuous collaborative development with our users.

Software Systems is a course taught by Dr. Allen Downey at Olin College of Engineering. It's an introduction to the design and implementation of system-level software, including operating systems, networks, and databases. Some of the topics that this course covers include processes and threads, memory and storage management, networking and inter-process communication, scheduling and synchronization.

During the spring semester of my junior year of college, I worked on a team of three students to implement an online chat room in C with a server that hosts multiple clients, which can speak to each other and play games together. Through this project, we figured out what web applications with multiple concurrent users need, such as sockets and data structures used to store user information. We also figured out how to identify potential memory leaks and fix them. Lastly, we picked up numerous useful tricks for using C, which we utilized to make our code more efficient and easier to read.

Once implemented, our online chat room allowed anyone to join and start sending messages if they were connected to the same network as other clients. Some additional features that we implemented included allowing clients to change their usernames, allowing two clients to play rock-paper-scissors against each other, allowing clients to privately message each other, and allowing clients to disconnect from the room at any time. In the end, when the chat room server is shut down, our program frees all memory allocations.

Because C is a programming language that was mostly new to all of the members of the team, we found this project to be very interesting and challenging. We were able to achieve all of our original learning goals, in addition to learning some additional skills along the way, as we became more familiar with using GLib hashtables, string operations in C, and more advanced usages of pointers. One of the most valuable experiences during this project was debugging and understanding the errors that we found.

The source code for this project is publicly available. If you're curious, you can take a look at it on GitHub here.

Complexity Science is a course taught by Dr. Allen Downey at Olin College of Engineering. It involves the study of complex systems and represents a new approach to science that investigates how relationships between parts give rise to the collective behaviors of a system and how the system interacts and forms relationships with its environment. Topics covered in this course include the structure and dynamics of complex networks; cellular automata, self-organized criticality, and generative systems; fitness landscape models of biological and technological evolution; and agent-based models of social and economic behavior.

During the fall semester of my junior year of college, I worked with another student to implement an agent-based model of a crime network based on a research paper that we studied. In our model, the agents are called Citizens and they inhabit a graph-like world called CrimeWorld, where each node is a street intersection. Each Citizen has several behaviors: they spend approximately 8 hours (the length of a work day) of their day moving around randomly in CrimeWorld, after which they return to their designated home and "sleep." They move randomly in one of the four cardinal directions for each time step. They can have two different roles: some Citizens are offenders (agents who commit crimes) and some are police (agents who prevent crimes). The offender agents share sets of characteristics, and for each time step, these characteristics help to determine whether they commit a crime or not.

After completing our initial model, my partner and I were interested in investigating what would happen if we added another factor that had not been present in our first model: the punishment of criminals who are caught. We therefore introduced a motivation characteristic for criminal agents and observe the results in our model.

The source code and final report for this project are publicly available. If you're curious, you can take a look at both of these on GitHub here.

During the spring semester of my sophomore year, I studied artificial intelligence independently with Oliver Steele as my advisor. One of my main objectives with this project was to maintain and improve my ability to use Java to solve problems, as at this point, I had not yet used Java in many technical projects. Another main objective was to learn how different AIs work and how to implement them algorithmically, as well as how to do so using Java.

My intention with this project was to create an entertaining game that is comprised of several different games. These other games included Go Fish, Battleship, and Tic-Tac-Toe. The overarching game is a Tamagotchi-styled game, in which the player must take care of an "Oliner" pet by playing games with it. Unlike Tamagotchis, though, the "Oliner" has a mind of its own and can intelligently keep up with the player in each game.

The AI for Go Fish is mostly hard-coded; the AI for Battleship makes use of a gradient map to determine where on the map a battleship is more likely to be located; and the AI for Tic-Tac-Toe uses the minimax algorithm and a generated game tree to produce its subsequent moves. I built a user interface for this project using Swing.

The source code for this project is publicly available. If you're curious, you can take a look at it on GitHub here.

Hacking the Library is a course that leverages the Olin Library and the areas of civic and community software as platforms on which to explore topics in software production, deployment, and operations. This is a studio course, with workshops, readings, and discussion around production development and operations as well as libraries. Topics are guided by the specific projects chosen. Projects are developed by individuals and teams and regular design reviews will help shape all dimensions of the projects so they may attain a maturity surpassing traditional software coursework.

In this unique course, I worked on three different webapps. The first webapp was a community calendar, the second webapp was a meditative space supplement, and the third webapp was a projects archive.

1. The Olin Community Calendar
The Olin Community Calendar was inspired by the fact that currently, email is the primary way for students to find out about events that have happened, are happening, or will happen on Olin's campus. For this project, I worked with another student to explore an alternative method of event storage and display. We came up with a community calendar, in which events are arranged by date and can be accessed and edited by the community.

The webapp itself is built with Flask, JavaScript, and HTML/CSS. There is a server that acts as the skeleton of the website, some static resources (including backgrounds and CSS), and several HTML templates that display data and other information to the user. Data are passed from HTML templates to the server via HTML form inputs. The Flask backend receives this information as a dictionary request form, and parses it to retrieve relevant information. Data is passed back into the HTML frontend via Flask. The events on the calendar are stored in a postgreSQL server provided by Heroku. Finally, psycopg2 is used as a connector between the Python code and the database.

The source code for this project is publicly available. If you're curious, you can take a look at it on GitHub here. In addition, the webapp is still hosted on Heroku and can be viewed here.

2. Calmmute
Calmmute, a project that I worked on with a partner, is inspired by the idea that commutes are cognitively important: having time between engagements allows our brains to finish processing the last thing we did before they begin working on the next. My partner and I noticed that people at Olin often either schedule meetings back to back, or use their awkward time between meetings to do more work. Both of these scenarios remove the precious commute time that can allow for stress relief and mental refreshment. As a result, we spun up Calmmute, an application that pairs simple, quick yoga exercises with time management to allow people to use their commute time more effectively.

Calmmute is built with Flask, JavaScript, and HTML/CSS. It uses a Python backend engine that takes user input to determine the type of yoga pose to display on its main page, then randomly selects an appropriate yoga pose. The main page displays the pose's instructions, as well as an instructional video.

The source code for this project is publicly available. If you're curious, you can take a look at it on GitHub here. In addition, the webapp is still hosted on Heroku and can be viewed here.

3. Project: Projects
I worked with a partner to survey students at Olin regarding how they handle recording all of the projects that they work on during their time in college. We found that 45 students at Olin (a whole 12.8% of the student body) confessed that recording projects is something of an ordeal. Some respondents created their own websites and portfolios, while about a third said that they had not created any kind of professional project portfolio or archive yet. In addition, about half of the respondents reported that they update their websites, portfolios, or PDFs only when job season is imminent. This all seemed to be a strong sign that the experience of recording projects at Olin has room for improvement. All of this feedback was gathered to form the basis for the creation of this project.

Project: Projects aims to make the lives of students easier by facilitating the creation of portfolios and lowering the activation energy for recording projects. It also aims to make the lives of faculty easier by placing all student projects in one convenient location. The frontend of the application is built using React, and the backend is built using Python and a MongoDB database.

At the end of the semester and the time of the final project freeze, my partner and I had created a webapp that allowed students to log on, search for their projects or other people's projects, and record new projects.

The source code for this project is publicly available. If you're curious, you can take a look at it on GitHub here.

Software Design is an introductory course that focuses on a model of computation as a set of simultaneous ongoing entities embedded in and interacting with a dynamic environment, for example: computation as it occurs in spreadsheets, video games, web applications, and robots. Throughout this course, students learn how to use Python to create games and tools through laboratory assignments, frequent homework assignments, and team projects.

During the spring semester of my freshman year of college, I worked with another student to create an augmented reality game based on Harry Potter. In this game, the player battles against Voldemort and his army of villains. Voldemort and other villains cast a number of spells, which inflict damage on the player. The player can return fire by waving their own wand in certain patterns, thereby casting spells.

This game was implemented in Python using a model-view-controller design pattern. We used OpenCV for the computer vision portion of the project, pygame for displaying the elements of the game, and various other libraries for implementing the rest of the game.

The source code for this project is publicly available. If you're curious, you can take a look at it on GitHub here.