Resume

Summary of Qualifications

Applied AI Researcher with 7 years of experience leading ML, NLP, and LLM research across healthcare, industry, and academic institutions.
Taught four university courses in data science, ML, and NLP, earning the 2023 Outstanding TA Award, and published 10 papers in leading venues (e.g., Empirical Methods in Natural Language Processing, Journal of Medical Internet Research).
Worked at top AI industry labs (Microsoft, Allen Institute for AI) and co-authored research with groups across several academic institutions (Weill Cornell, University of Washington, University of Illinois Urbana-Champaign).

Ithaca, NY | Aug 2021 – Present

AI Evaluation: Built an LLM-based system for radiology report evaluation with Microsoft and engineered an LLM-as-a-judge evaluation framework increasing expert agreement by 24%.
AI Implementation: Leveraged model fine-tuning and few-shot learning to identify unmet user needs in endometriosis online communities, as well as perceptions of ablation and excision surgeries.
Causal Inference: Designed a randomized experiment with 3,000 participants to model their reading preferences.
Published work in Journal of Medical Internet Research (JMIR), Journal of Minimally Invasive Gynecology, and Empirical Methods in Natural Language Processing (EMNLP) Findings 2026.

Redmond, WA | May 2026 – Aug 2026

Built a production pipeline that automatically fine-tunes the nursing AI scribe to hospital-specific needs via prompt optimization.
Improved nursing documentation accuracy by up to 44 percentage points using the AI-optimized prompts.
Presented results to the HLS organization, including the Partner Applied Science Manager and Corporate VP.

Seattle, WA | Jun 2025 – Apr 2026

Developed a query reformulation system that increases the recall of scientific information in the output of the Asta AI research assistant by 50%.
Built two automatic evaluation pipelines leveraging expert-curated and augmented user data to iteratively test and improve model performance against key metrics.
Designed a user study to measure the usefulness and relevance of the system’s suggested query reformulations.

New York City, NY | Jun 2023 – Aug 2023 | Funded by the Siegel Family Endowment PiTech PhD Impact Fellowship

Identified three alerts frequently overridden by nurses by analyzing around 3,000,000 clinical interruptive alerts.
Used data analysis and statistical testing to demonstrate that revising alert criteria can reduce unnecessary alerts by up to 94%, while improving alert design can decrease overridden alerts by up to 64%.
Prompted hospital leadership to implement organization-wide routine alert revisions by presenting findings to administration and informatics experts.
Published work in AMIA 2025.

Bologna, IT | Aug 2020 – Jul 2021

Combined network analysis and machine learning to demonstrate that the strength of citational ties between candidates and evaluating commissions affects academic hiring outcomes.
Revealed that open bibliographic datasets provide lower coverage of publications in the humanities and social sciences compared to STEM by building a pipeline to query and collect 275M+ bibliographic and citation records.
Published in Quantitative Science Studies and Scientometrics.

Aarhus, DK | Sep 2019 – Dec 2019

Applied statistical analysis and natural language processing to demonstrate the lack of urban spaces in sci-fi novels compared to general fiction across 16,000+ novels.
Discovered that the presence of urban spaces in sci-fi is high at the beginning of the 20th century, declines in the 30s and 40s, and increases in the 50s.
Published work in Journal of Cultural Analytics.
Awarded the Research Abroad Scholarship by the University of Bologna.

Ithaca, NY | Aug 2021 – Present

Taught several courses: Introduction to Data Science, Studying Social Inequality Using Data Science, Text Mining History and Literature, Books as Data.
Led weekly sections on data science, machine learning (ML), and natural language processing (NLP).
Delivered lectures on social media analysis with NLP and statistical techniques for causal effect estimation.
Curated class activities on preprocessing large raw census data and applying ML to predict social outcomes.
Received the 2023 Outstanding TA Award.

Cornell University — Ph.D. in Information Science | Ithaca, NY | 2026 GPA: 4.00 / 4.00 · MSR Azure Cloud Computing Grant ($18,700)

Cornell University — M.S. in Information Science | Ithaca, NY | 2024 GPA: 4.00 / 4.00

University of Bologna — M.A. in Comparative Literature | Bologna, IT | 2020 GPA: 110 / 110 cum laude

University of Bologna — B.A. in Foreign Linguistics and Literature | Bologna, IT | 2017 GPA: 110 / 110 cum laude

Selected 5 out of 10 total publications, 8 as first author.

Federica Bologna, Jean-Philippe Corbeil, Matthew Wilkens, Asma Ben Abacha. VERT: Reliable LLM Judges for Radiology Report Evaluation. Submitted to ACL ARR, 2026.
Federica Bologna, Tiffany Pan, Matthew Wilkens, Yue Guo, Lucy Lu Wang. CQA-Eval: Designing Reliable Evaluations of Long-Form Clinical QA under Resource Constraints. EMNLP Findings, 2026.
Federica Bologna, Rosamond Thalken, Kristen Pepin, Matthew Wilkens. Endometriosis Online Communities: Quantitative Analysis. Journal of Medical Internet Research, 2025.
Federica Bologna, Anand Reddy, David Silvestri, Michael Bouton. Revising BPA triggers and inclusion criteria helps reduce nurses’ fatigue. Poster at AMIA, 2025.
Federica Bologna, Ian Lundberg, Matthew Wilkens. Causal Effect of Character Gender on Readers’ Preferences. Computational Humanities Research, 2025.

Programming Languages: Python, R, Bash, SQL
Methods: Machine Learning, Large Language Models, Fine-Tuning, Optimization, Natural Language Processing, Data Science
Tools: PyTorch, HuggingFace
Interests: Poetry fanatic (Meadowlands is a favorite), aspiring gym rat, anime binge-watcher (Frieren at the moment)