Gabriel Agostini smiling, with Scotty the black-and-white cat perched on one shoulder
CV

Gabriel Agostini

PhD student in Information Science

Cornell Tech, Cornell University

About

I am a fifth-year PhD student in Information Science at Cornell Tech, working with Emma Pierson and Nikhil Garg.

My research leverages spatial machine learning methods and creates novel datasets to inform more equitable urban policies. I focus on addressing challenges related to sparse and biased spatial data: specifically, how to transform coarse, crowdsourced, and irregularly collected information into actionable insights for improved city resource allocation.

Currently, I work with the New York City Department of Transportation to evaluate bike lane usage and bike ridership across the city. In recent work, I examined under-reporting bias in resident crowdsourcing systems (such as New York City’s 311), migration patterns in the United States at scale, and disparate access to urban amenities. I rely on spatial analysis techniques combined with computational and statistical frameworks including Bayesian optimization, hierarchical modeling, and network analysis. I also co-organize a working group on Urban Data Science under EAAMO Bridges, and we are always looking for members or guest speakers⁠—let me know if you would like to participate!

Prior to joining Cornell, I earned a B.S. in Applied Mathematics and a B.A. in Urban Studies from Columbia University. I have written quite a lot about cities back then, and included some of this work under the fun tab.

Send me an email if you are interested in my research! Here is a copy of my CV.

News

  1. I wrote a small reflection about my (really fun!!!) experience producing a short film about urban data with the amazing Tianyi Sun. Looking forward to sharing more soon! Figure 2.3 was peer reviewed already.

  2. Our paper on bike route choice with large-scale e-bike GPS data will appear at EAAMO 2026 (non-archival). Reach out if you’ll be there!

  3. MIGRATE received the IPUMS Spatial Student Research Award.

  4. I will be giving a talk at Penn State’s EEPI Seminar on April 1st!

  5. MIGRATE received the IGIF Student Paper Award from the American Association of Geographers! If you will be in SF for the AAG annual meeting, let’s chat!

Older newsHide older news (3)

  1. Inferring fine-grained migration patterns across the United States is out in Nature Communications!

  2. I wrote a blog post about my work with the NYC DOT on map-matching GPS pings for Citi Bike trips. We are now building a route choice model to understand cycling behavior in the city—stay tuned!

  3. I joined the NYC DOT Cycling & Micromobility unit as a fellow. Excited to work in city government, contributing to better urban mobility infrastructure!

Publications

* denotes equal contribution

Click on a title to read the paper (when a draft is available).

Working papers

  1. 2026

    Modeling Biker Preferences with Large-Scale E-Bike GPS Data

    Gabriel Agostini, Nikhil Garg, Emma Pierson

    EAAMO 2026 (non-archival)

  2. Quantifying Spatiotemporal Gaps in Accessibility with Fine-Grained, Open-Source Amenity Datasets

    Gabriel Agostini, Zhixuan Qi, Katelyn Liu, Shreya Sriram, Esther Li, Emma Pierson, Nikhil Garg

  3. The Forest from the Trees: Lessons on Public-Sector Deployment from an Academic Collaboration with NYC Department of Parks and Recreation

    Erica Chiang, Katelyn Liu, Anum Ahmad, Zhi Liu, Gabriel Agostini, Tyler Gibson, Uma Bhandaram, Nikhil Garg

Publications

  1. 2026

    Inferring fine-grained migration patterns across the United States

    Gabriel Agostini, Rachel Young, Maria Fitzpatrick, Nikhil Garg, Emma Pierson

    Nature Communications 17, 1265

    • IPUMS Spatial Student Research Award
    • AAG-IGIF Student Paper Award
  2. 2025

    Inequalities in Mobility Patterns: Reconciling Access and Travel in American Cities

    Talia Kaufmann*, Gabriel Agostini*, Trivik Verma, Dan O'Brien

    19th International Conference on Computational Urban Planning and Urban Management (CUPUM), pp. 1–12

  3. 2024

    Topics, Authors, and Institutions in Large Language Model Research: Trends from 17K arXiv Papers

    Rajiv Movva*, Sidhika Balachandar*, Kenny Peng*, Gabriel Agostini*, Nikhil Garg, Emma Pierson

    NAACL 2024, pp. 1223–1243

  4. 2024

    A Bayesian Spatial Model to Correct Under-Reporting in Urban Crowdsourcing

    Gabriel Agostini, Emma Pierson, Nikhil Garg

    Proceedings of the AAAI Conference on Artificial Intelligence 38, pp. 21888–21896

    • Oral presentation
  5. 2022

    Identifying Urban Morphology from Street Networks with Graphlet Analysis

    Gabriel Agostini, Juliana Gonçalves, Trivik Verma

    30th Annual Geographical Information Science Research UK (GISRUK), pp. 1–6

  6. 2021

    A Category-Theory Approach to Construction Ontologies in Subsurface Mass Transit

    Gabriel Agostini, Yushan Zhang, Debra Laefer

    International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLVI-4/W4, pp. 125–130

Talks

Invited talks & lectures

  1. Penn State University

    EEPI Seminar

  2. New York University

    Urban Lab: Immigration and Housing in Queens

  3. Vision Zero Initiative

    NYC Research on the Road

  4. Barnard College, Center for Computational Science

    CSC Workshop

  5. Cornell Tech

    Urban Systems

Older talksHide older talks (3)

  1. Cornell Tech

    Applied Data Science

  2. Cornell Tech

    Urban Data

  3. Cornell Tech

    Urban Data

Conference talks

  1. Inferring Fine-Grained Migration Flows across the United States

    AAG Annual Meeting, San Francisco

  2. Inequalities in Mobility Patterns: Reconciling Access and Travel in American Cities

    CUPUM 2025, London

  3. A Bayesian Spatial Model to Correct Under-Reporting in Urban Crowdsourcing

    The 38th Annual AAAI Conference on Artificial Intelligence, Vancouver

  4. Identifying Urban Morphology from Street Networks with Graphlet Analysis

    30th Annual Geographical Information Science Research UK (GISRUK) video

  5. A Category-Theory Approach to Construction Ontologies in Subsurface Mass Transit

    3D GeoInfo Conference

Older talksHide older talks (2)

  1. Operationalizing Equity in the San Francisco Student School Assignment

    Stanford Data Science for Social Good video

  2. Knot Surgery and Integer Characterizing Slopes

    Columbia University Mathematics Department video