I am excited about: generative models, technology for writing, & the ethics of data work.
I am a human-AI interaction researcher, with a focus on technology for impactful writing and understanding the limits and capabilities of large language models. I'm currently a Postdoctoral Fellow in Computer Science with Elena Glassman at Harvard University. I recently completed my PhD at Columbia University with Lydia Chilton. I have been supported by an NSF Graduate Research Fellowship and the Brown Institue for Media Innovation.
I'm also a poet and essayist. I write about the ocean, our looming climate crisis, my nieces and nephews, and the intersection and distinction of the arts and science. I'm working on a poetry manuscript called 'whalefall'. In 2020 I was named a Brooklyn Poets Fellow and was a 2020-2021 CultureHub Resident Contributing Writer. I was a Writer in Residence at the Vermont Studio Center in 2022. I'm currently on the editorial team for taper, a literary magazine for small computational pieces.
I used to build machines and sensors. I hold a Bachelors of Science in mechanical engineering from MIT, where I recieved the Carl G. Sontheimer Prize for Excellence in Innovation and Creativity for my work on soft robotics. I worked at two startups: Rest Devices (we built baby monitors) and Soofa (we built smart city furniture).
See my CV, Semantic Scholar, or Google Scholar for a full list of publications.
My thesis asks how writers use computer-generated text. In particular I look at constrained, creative writing tasks, like metaphors and science writing. I've developed new interfaces and technologies that generate text for writers, as well as studied existing systems. My thesis culminated in a study that asks when and why writers turn to computers versus humans for support.
Select publications:
As large language models have begun to swallow up research on text generation, I've become increasingly concerned with their problems. I'm interested in developing tools that give end-users the ability to understand bias in text generation for their particular use case, as well as foundational work on using smaller models with well-understood datasets.
Select publications:
During my PhD, I was always looking for people's materials (thesis proposal, fellowship application essays, etc.) to use as examples, particularly anyone whose thesis was similar to mine either in content/subject area or form. Here are my own materials, for those looking for examples:
Application for NSF Graduate Research Fellowship (2017)
[personal essay] [research plan]
Application to Columbia University PhD program (2017)
[statement of purpose]
Candidacy (2020)
[list of papers] [presentation]
Thesis Proposal (2021)
[original document] [revision]
Dissertation (2022)
[dissertation] [presentation]
I have created some freely available writing tools. These tools range from a simple website that will rearrange the lines of your poem, to custom generated thesauruses based on different styles and topics. Poetry is often about seeing the familiar anew, and computation provides a way to interact with your own writing in curious and foreign ways.
My poems, essays, and small computational pieces have been published in a variety of venues, as well as been performed live. I'm interested in computation as a way to deepen how we can experience literature, but I'm also invested in writing as a manual, human process. When computation enters my work, I prioritize the beauty of language over the concept of the algorithm, and in this way I often feel more aligned with the concrete poets than the computational ones. I love the browser as a place of play, and see websites as a medium worth taking seriously.
I curate a list of publications that accept computational poetry.
Last updated August, 2023