Hi there! I'm Tiffany!

MIT '22/'23 | Software Engineer @ Pinterest Labs | NYC

tiffany's profile picture

Interesting factoids about me:

  • I am a current MEng student with the HCIE group at CSAIL. As an undergrad, I studied computer science and humanities and mathematics at MIT.
  • My research interests lie in the intersection of human-computer interaction and artificial intelligence to help people unleash their creativity and understand writing. My current master's thesis hinges on making automatically illustrating poetry using generative AI.
  • I'm an incoming software engineer at Pinterest Labs! I'll be working in their visual search team. building technologies on top of our ML models to help Pinners find similar images to their boards.
  • At MIT, I was part of the THINK committee, where I select several research project proposals from high school students to mentor and fund their research. I also was the head teaching assistant for 6.1200 (Math for Computer Science).
  • In my free time, I write short stories and poetry. I also love traveling and going for hour-long walks around the city.

Highlighted Work Experience

Jump Trading

Intern | Jun '22 - Aug '22 | Chicago, IL

I worked as a software engineering intern at Jump Trading on their UI team. There, I visualized risk exceptions in a live chart, showing when teams are potentially overexposing the firm, and reduced the memory usage in-browser for the corresponding web application.

Meta

Intern | Jun '21 - Aug '21 | San Francisco, CA (remote)

I worked as a software engineering intern at Meta. There, I built and analyzed several UI components to raise awareness of the onsite checkout feed in the Shops tab, resulting in more engagement with the feed.

MIT Department of Electrical Engineering and Computer Science

Head Teaching Assistant | Feb '19 - Present | Cambridge, MA

I am a current teaching assistant for the Mathematics of Computer Science (6.1200) class at MIT. I was also a lab assitant for the Fundamentals of Programming (6.009) class at MIT.

Microsoft Research - Semantic Machines

Microintern | Jan '21 - Feb '21 | Cambridge, MA (remote)

I worked as a software engineering microintern at Microsoft Research - Semantic Machines. There, I built a decision tree to predict the validity of crowdsourced data for computer-generated code.

Interested in my resume? Here it is!

Research

Closer Worlds: Using Generative AI to Facilitate Intimate Conversations

Tiffany Chen*, Cassandra Lee*, Jessica R Mindel*, Neska Elhaouij, Rosalind Picard (* equal contribution)

CHI 2023

Deep emotional intimacy is a foundational aspect of strong relationships, but the digital tools we use to communicate often limit rather than empower our feelings of connection. Two compelling strategies technologists have used to counteract such trends include games and generative AI art. In this paper, we design and test Closer Worlds, an ML-assisted 2-person game that fosters emotionally intimate conversations through co-creative world-building. We explore design principles inspired by facilitation methods and assess their effectiveness in a pilot study. We find that Closer Worlds elicits some self-disclosure behavior, but less than a social game without generative AI. However, participants clearly enjoy the unique affordances offered by visualizing shared values, which suggests that this method offers a comfortable and novel avenue for meaningful conversations. We conclude by discussing future ways in which co-creative games might leverage generative techniques to foster pro-social environments.

Kaleidoscope: Semantically-grounded, context-specific ML model evaluation

Harini Suresh, Divya Shanmugam, Tiffany Chen, Annie G Bryan, Alexander D'Amour, John Guttag, Arvind Satyanarayan

CHI 2023

Desired model behavior often differs across contexts (e.g., different geographies, communities, or institutions), but there is little infrastructure to facilitate context-specific evaluations key to deployment decisions and building trust. Here, we present Kaleidoscope, a system for evaluating models in terms of user-driven, domain-relevant concepts. Kaleidoscope’s iterative workflow enables generalizing from a few examples into a larger, diverse set representing an important concept. These example sets can be used to test model outputs or shifts in model behavior in semantically-meaningful ways. For instance, we might construct a “xenophobic comments” set and test that its examples are more likely to be flagged by a content moderation model than a “civil discussion” set. To evaluate Kaleidoscope, we compare it against template- and DSL-based grouping methods, and conduct a usability study with 13 Reddit users testing a content moderation model. We find that Kaleidoscope facilitates iterative, exploratory hypothesis testing across diverse, conceptually-meaningful example sets.

Projects

Google StreetBuddy

My team and I built a multimodal Google street view, allowing for users to traverse through Google Street View through gesture (read through a Leap Motion) and through voice.

python, typescript |


Portable violin

My team and I built a violin on a glove. I specifically worked on building the main glove and refining the code.

arduino


TechX application viewer

My team and I developed an interface to allow committee members to easily read applications for THINK and XFair (a conference at mit) and accept/reject applicants.

react, mongodb

application viewer project picture

Stay Notified Cambridge

My team and I made an interface for the Cambridge City Council to notify its constituents about meetings being held.

vue

application viewer project picture

Virtual activities midway website

I helped TechX make a virtual activies midway interface so that clubs and incoming freshmen at MIT can link together, specifically the filter function for the different clubs.

react

activities midway interface picture

FearlesslyGirl app

I helped FearlesslyGirl, an organization dedicated to inspiring girls to take on the world confidently and fearlessly, to build the hangout section of their new app.

react-native |


THINK website

I redesigned the front-end of the MIT THINK website to look more modern and sleek.

angular, bootstrap


Assassin game organizer

I worked with a team to develop an assassin's game organizer. A user can register and start/join existing assassin games. The app does all administrative parts of the game, such as randomly assigning targets and administering a way to mark deaths.

flask, javascript, bootstrap, sql databases |


Mario Party online game

I worked with a team to develop an online version of Mario Party. Two players can play simultaneously, going across the board and playing minigames to earn/lose coins. Players can also continue old games from the home page.

flask, javascript, foundation, sql databases |


Safari Zone game

My partner and I developed a game in processing simulating the safari zone minigame found in many main Pokemon games. Players can choose their character and catch pokemon, collecting them all in several safari zone habitats.

processing |


Spell checker

My partner and I developed a spell checker. By inputting text, the program checks each word with the dictionary and, if the word is not found, spell checks it. the user can also respell-check words.

java |

spell checker project picture

Want to see more? Check out my github.

contact me

tiffc [at] mit.edu

© tiffany chen, 2023