Hi! This is Sonia Castelo Quispe. I am a research engineer and PhD student in Computer Science at New York University, working with the Visualization, Imaging and Data Analytics (ViDA) Lab. My research focuses on visual analytics, human-computer interaction, machine learning, and augmented reality for AI-assisted task guidance and data-driven model analysis.

Research

My work focuses on interactive visualization, data exploration and curation, and model analysis, with an emphasis on developing novel visual analytics approaches and systems for understanding complex data and supporting real-world, data-driven tasks.

Current and recent projects include OSCUR (NSF), DARPA PTG, D3M, MEMEX, Vizier, ARGUS, and HuBar. My research has led to publications in VIS, CHI, IEEE VR, TVCG, VLDB, SIGMOD, CIDR, and related venues.

Selected research projects

• OSCUR: Open-Source Cyberinfrastructure for Urban Computing
  National Science Foundation (2024–2029) Project Website Hugging Face Press
  Developing comprehensive tools for urban data collection and analysis, addressing climate change, walkability, bikeability, and urban accessibility.

• OpenSpace: Interactive Visualization of the Known Universe
  NASA Science Mission Directorate (2017-2025) Project Website GitHub
  OpenSpace is open-source interactive data visualization software designed to visualize the entire known universe and portray our ongoing efforts to investigate the cosmos.

• PTG: Perceptually-enabled Task Guidance
  DARPA (2021–2024) DARPA Program NYU Press Hubar Press ARGUS Paper TIM Paper
  Building AI-assisted guidance systems for complex physical tasks using AR headsets, wearable sensors, and real-time feedback.

• D3M: Data Driven Discovery of Models
  DARPA (2018–2021) DARPA Program Project Website
  Creating automated model discovery systems to support subject matter experts without formal data science training.

• Vizier: Collaborative Streamlined Data Curation
  National Science Foundation (2018–2021) Project Website
  Unifying curation and data exploration through provenance, with a notebook interface, spreadsheet-style views, and workflow tracking.

• MEMEX: Fighting Against Illegal Activities
  DARPA (2016–2018) Project Website Press Third place at NYU Research Expo
  Locating, analyzing, and exploring hard-to-find information on the web surface and the deep web.

• E-share Miner: Information Management Supported by Knowledge Discovery through Topic Taxonomy
  Itera - Inovação e Desenvolvimento Tecnológico (2015–2016)
  Discovering new knowledge through topic taxonomy in juridical text collections using text mining and information visualization.

• Earlier research projects at UNSA / CONCYTEC — Arequipa, Peru
  Computer vision and automation projects including scallop selection and processing, brazil-nut classification, and mechatronic assistance systems.

How these projects connect

Across these projects, my work has focused on interactive visualization, human-computer interaction, machine learning and AutoML, augmented reality for task guidance, text mining and information retrieval, data curation and provenance, and computer vision and image analysis.