Last Fall, the CDH Research Software Engineering team traveled near and far to share their cutting-edge work and learn from others about new developments in the fields of research software engineering and digital humanities.

🔔 Philadelphia

From October 6–8, the RSE team participated in the third annual conference of the United States Research Software Engineer Association. Held in Philadelphia, US-RSE’25 brought together RSEs from across universities, laboratories, industry, and other institutions, as well as their managers and allies, to discuss “Code, Practices, and People.” CDH represented the small — yet mighty! — Humanities contingent in a field of mostly scientists and social scientists.

Lead RSE Rebecca Sutton Koeser presented a notebook on undate, a Python package for computing with uncertain and partially-unknown dates, such as those in Sylvia Beach’s lending library records and in the fragmentary Hebrew texts in the Princeton Geniza Project. Koeser also collaborated on two posters: “Community Code Review in the Digital Humanities,” which detailed the history, process, and future work of the DHTech Code Review Working Group; and “Surveying the Digital Humanities Research Software Engineering Landscape,” which reported on survey results about the backgrounds and career paths of DH developers.

Postdoctoral Researcher Christine Roughan presented a paper, also co-written with Koeser, titled “Integrating ATR Software with University HPC Infrastructure: balancing diverse compute needs.” The paper and corresponding presentation described the methods and outcomes of Bringing HTR to the HPC: A Pilot to Customize eScriptorium for Princeton, a Research Partnership with the CDH under the umbrella of the Princeton Open HTR Initiative (funded by a 2024–25 Princeton Language + Intelligence Seed Grant). Conference attendees were fascinated to hear about how Koeser and Roughan implemented an instance of eScriptorium — the current leader in open-source handwritten text recognition software — on Princeton’s high-performance computing hardware, which enabled professors and students without advanced technical skills to train large text-recognition models customized to their documents’ needs.

Assistant Director Jeri Wieringa and Project Manager Mary Naydan presented on a panel about supporting and managing RSE projects. Their presentation “Creating Research Software with Humanities Faculty” highlighted the CDH’s chartering process, which helps transition humanities faculty from the individual, expansive mode of traditional humanities scholarship to the collaborative, modular mode of computational research. Their illustrative opening skit, which set the stage for the talk, drew lots of laughter and resonated with audience members. The rest of the panel was just as engaging, sharing lessons learned from many different types of organizations and fields, from the multi-institutional development of medical technologies, to a large laboratory focused on national security, to a lone RSE’s personal project management workflow at a research university (Naydan’s favorite presentation of the conference!).

The technical talks affirmed that the CDH RSE team is ahead of the curve on best practices in Python development, such as using uv for installing packages and choosing Marimo over Jupyter for notebooks. Koeser noted the field-wide shift in starting to think about notebooks as a form of publication, and CDH Research Software Engineer Hao Tan was inspired by Reed Maxwell’s keynote on creating groundwater simulations using physics-informed machine learning. Tan reflects, “Explainability is still a real issue, but rather than rejecting AI outright, we should learn to leverage its strengths and mitigate its weaknesses — through comparative evaluation, transparent step-by-step reasoning, and other methods we develop.”

Many of the conference’s presentations — from Maxwell’s keynote to the Birds of a Feather workshop “AI in Practice” — showed the field of research software engineering grappling with, adapting to, and incorporating AI. While our team entered the conference thinking our challenges were somewhat unique to the humanities, we were surprised to see RSEs from across disciplines encountering similar challenges around this topic: from defining research questions, to gathering sufficient data, to disabusing researchers about what AI can actually do.

🇱🇺 Luxembourg

Unsurprisingly, AI was also a popular topic at the 2025 Computational Humanities Research Conference, held at the Luxembourg Centre for Contemporary and Digital History (C²DH) at the University of Luxembourg from December 9–12, 2025. Many of the presentations focused on benchmarking various models for automatic transcription tasks, or using chatbots to scale up annotation data, from identifying “acts of God” in contemporary Christian fiction to assessing a popular song’s “narrativity.” Miguel Escobar Varela’s keynote, “‘A watch by Kran Kamu’: Exploratory fine tuning for cultural reliability,” discussed using supervised fine-tuning on large open-weight models to yield reliable results within highly specific cultural contexts, such as Southeast Asian historical newspapers — a common problem facing computational humanities researchers given the specialized nature of our data and the scarcity of it for fine-tuning.

Rebecca Koeser and Mary Naydan presented a poster based on their short paper “Unstable Data and the Unusual Case of the Prosody Excerpt in the Digital Library” (co-authored with Meredith Martin). Using the HathiTrust materials contained in the Princeton Prosody Archive as a case study, Koeser and Naydan cautioned researchers that the page-level data provided by cultural heritage aggregators is not as stable as we might assume. This instability can lead to erroneous data, flawed conclusions, and difficulties building on previous scholarship.

Hao Tan delivered a lightning talk, “When Larger LLMs Aren’t Enough: Word Segmentation in Historical Chinese Texts,” which used word segmentation in historical Chinese texts as a case study to highlight how large language models, while powerful, can quietly introduce risks when applied to humanities research. The talk sparked conversations with researchers working on East Asian materials across Europe, the US, and Singapore, especially around the tricky parts of historical text processing, with projects ranging from power relations in historical fiction to poetic imagery and stylistic change in epitaphs.

In addition to Tan’s lightning talk, Koeser and Naydan found two others particularly interesting: Katarina Mohar’s on “Speculative Reconstruction and the Ethics of the Fragment: Early Experiments with Generative AI in Art History,” and Antonina Martynenko, Artjoms Šeļa and Petr Plecháč’s on "Where Empires End: Tracing the Geography of a ‘Soaring Spirit’ in Poetry.” Mohar discussed the possibilities and limitations of using Generative AI to fill in gaps in medieval paintings, and provided practical recommendations for how to use it responsibly. Martynenko et al. examined the spatial imagination of European poets by mapping the distance and directionality of place mentions.

For Koeser, one of the most interesting presentations was “Cluster Ambiguity in Networks as Substantive Knowledge,” which describes a method for running a clustering algorithm multiple times to measure how often edge nodes connect nodes in the same community, allowing researchers to identify ambiguous data. Koeser is interested in the interpretive power and potential applications of this method, such as identifying ambiguous characters in novels. Another highlight was Taylor Arnold and Lauren Tilton’s presentation on “Sitcom Form and Function: Pacing and Production in a Collection of Thirty U.S. Series,” which examined how trends in visual and aural pacing changed over time using a combination of large-scale computational analysis and close reading: an example of truly multimodal research and scalable reading.

The RSE team was energized by these shifts in the field: leaning into ambiguity; using audio, video, and visual data rather than defaulting to text; and combining different scales of reading (close and distant) to draw more responsible conclusions. We are excited to carry what we learned into our work this year on multilingual machine translation and term clustering in music theoretical texts and using vision-based LLMs to aid historical document transcription and data extraction.

Back in January 2025, Cecily Swanson, ODOC’s Associate Dean for Academic Advising since 2020†, attended my Project Management 101 Wintersession. She found the workshop so helpful that she wanted her colleagues to experience it, too, before the start of the new academic year.

The workshop covered everything from project management theory and methods to project charters to best practices for running effective meetings. While the CDH specializes in project management for software engineering within a humanities context, the ideas apply to a variety of fields, from student advising to event planning. The following slide shows participants’ responses about the concepts they were most excited to adapt:

Swanson reported that the workshop was “a revelation for our team of academic deans,” who “left the workshop equipped to identify challenges in our current projects and enact solutions.” She shared, “The workshop gave us concrete strategies for improving the core features of our work.”

My highlights included playing the penny game — a timed interactive game in which participants flip and pass pennies to learn about the Agile mindset — and having a deep discussion with the group about which project management skills are the most essential. Part of the fun in presenting these concepts to professionals in different academic fields is that I get to see familiar concepts in a new light. I’m always excited to hear the insightful questions and comments that this material generates.

Swanson summed up the pragmatic merits of the CDH’s project management outreach to students, faculty, and staff on campus: “We all know the value of collaboration, communication, and change management for successful project management; Naydan’s workshop offers the tools to actually put these values into practice.”

†Swanson has been newly appointed as the associate dean for undergraduate affairs for the School of Engineering and Applied Science, and we wish her well in her new role!

What does it mean to be human in the age of artificial intelligence? As emerging technologies reshape daily life, DH Project Manager Mary Naydan *23 (English) turned to literature for answers. Her first-year seminar, Speculative Fiction: From Pygmalion to ChatGPT (FRS 142), examined how imaginative storytelling in science fiction, fantasy, and horror has long anticipated today’s AI debates.

The course began with classics like Mary Shelley’s Frankenstein (1818), prompting students to consider the ethical responsibilities of creators and ask, “Who should be held liable when a creation causes harm?” Despite its age, Shelley’s novel remains strikingly relevant in discussions of today’s rapidly evolving and largely unregulated AI landscape.

In the same week, students studied the origins of the “Frankenstein Complex,” a term coined by I, Robot author Isaac Asimov to describe fears of AI turning against its creators. They examined how such anxieties shape not only fiction but also real-world discourse, including a 2023 open letter from tech leaders calling for a pause in AI development.

By tracing concerns about artificial intelligence from Ovid’s account of the Pygmalion myth (8 CE) to contemporary works by authors Ted Chiang and Nnedi Okorafor, the course equipped students with the tools to think critically about AI. It empowered them to engage with the technology thoughtfully and be aware of both its possibilities and its ethical implications.

“Today’s students are inundated with AI tools, whether they realize it or not… and it is getting harder and harder to opt out,” says Naydan. Her goal is to help students make informed choices about the role of AI in their lives, as students and beyond.

This spring’s agenda was immersive and certainly one to remember.

Fostering AI literacy through critical engagement

Students built AI literacy through a series of hands-on labs designed to demystify how these systems work and encourage critical reflection. In the first lab, they used Google’s Teachable Machine to train a simple computer vision model to recognize yoga poses, observing firsthand how supervised machine learning relies on data patterns rather than human-like understanding. For example, a model trained only on right-legged tree poses struggled to recognize the same pose with the left leg, revealing the limits of generalization and challenging the tendency to anthropomorphize AI. Subsequent labs explored AI bias and emotional mimicry.

Students also honed their prompt engineering skills by tasking ChatGPT with generating a hypothetical tenth story for I, Robot, then critically evaluating the outcome. In another session, they explored Sudowrite—an AI tool designed for creative writers—to examine the potential advantages and limitations of AI-assisted storytelling.

Later in the semester, during a unit on climate, students visited Princeton’s High Performance Computing Research Center (HPCRC), which houses the campus’s AI research infrastructure. The trip offered a concrete sense of AI’s environmental footprint and how Princeton’s graphics processing units (GPUs) compare to the scale and impact of tech giants like Meta, Tesla, and Google. As a LEED-certified facility, HPCRC demonstrates one model of sustainable infrastructure, and it left an impression, not least because students were fascinated by its eco-friendly lawn care: sheep to mow the grass!

“Visiting the HPCRC was a fascinating experience that deepened my understanding of high-performance computing and its role in cutting-edge research,” says Chinmayi Ramasubramanian ’28. “I have often used the Adroit computing cluster in class, and coming in, I was excited to see the actual hardware that powered my computations.”

In preparation, students read Kate Crawford’s Atlas of AI chapter “Earth,” which challenges the illusion of a “green” tech industry. They learned that AI systems rely on massive computational power and energy, often hidden behind the metaphor of “the cloud.” The visit, paired with discussion, helped students connect abstract environmental critiques to the tangible materiality of AI infrastructure and to consider how AI’s growth drives both energy consumption and resource extraction globally.

“Seeing the physical hardware up close and learning about how everything runs made the complexity of these systems feel so much more real,” states Quinn Challenger ’28.

Discussing AI with award-winning fiction writers

A highlight of the course was the opportunity for students to engage directly with two of today’s most influential speculative fiction authors, Ted Chiang and Nnedi Okorafor, both of whom visited the class before giving public lectures as part of CDH’s Humanities for AI series. Naydan emphasized the value of these visits: “It is one thing to talk abstractly about the relationship between AI and creative writing; it is another to hear from creatives directly—what they think about AI, how they use it (or choose not to), and why.”

To prepare for the visits, students read and discussed each author’s work. Chiang’s work (The Lifecycle of Software Objects and nonfiction essays in the New Yorker) prompted conversations about the ethical implications of developing AI within capitalist systems and ChatGPT as an impediment to developing “cognitive fitness” in college. Okorafor’s work (Death of the Author and “Abracadabra”) inspired discussion on AI’s positive potential to support people with disabilities and improve healthcare.

“There was an electric atmosphere to the conversations,” Naydan recalls. “I was so impressed by the students’ willingness to think deeply, take risks, and have fun with ideas.”

The visits left meaningful impressions on the students.

“Being a student-athlete, [Okarafor's] story about her experience as an elite athlete who turned a career-ending injury into a path toward writing resonated with me,” says Nathan Banos ’28. “It reminded me that it’s not about what happens to you, but how you respond. Their visits to our classroom were the perfect way to immerse ourselves in the world of both speculative fiction and AI, creating a unique learning environment and lasting memory for me.”

Stephanie Ko ’28 enjoyed the thoughtful balance in the course’s exploration of AI’s technological foundations and its human-centeredness, to which the visits added depth. “I was honestly awed to see how Ted Chiang simply sees and fulfills the need to be a well-informed member of the perpetual discussion regarding the future of AI,” she says, “which is a trait that I think many of us will aspire to develop and carry with us through our education at Princeton.” Of Okorafor, she shares, “She was incredibly transparent that her hopes for AI stem from her personal struggles and frustrations, and I think this perspective was a refreshing reminder that the development of AI and the role we allow it to have is a fundamentally human problem.”

What comes next?

“If my students take just one thing away from my class, I hope it is the idea that AI is not objective, perfect, or neutral,” says Naydan. “It is fallible, because it’s only as good as its training data; it encodes and perpetuates human biases; and it is complicated in how it can help and hurt humanity, often simultaneously.”

She also stated that by the end of the course, there was no single answer to how literature responds to technological change in our society. Instead, students encountered a range of creative approaches: Okorafor imagines the technologies she hopes society will build; Kai-Fu Lee and Chen Qiufan use “scientific realism” to depict core concepts of machine learning; Terence Taylor explores the exploitation of labor in AI-driven workplaces through horror; and Philip K. Dick uses anthropomorphized AI to probe deeper moral and existential questions about what it means to be human.

Ultimately, the course underscored that literature has long been a tool for making sense of the world. As AI continues to reshape society, storytelling not only helps us understand emerging realities—it also offers a means to imagine and influence what comes next.