Networks, science, and AI across disciplines.

Trained as a physicist, I am an interdisciplinary researcher working across network science, science of science, artificial intelligence, and visual computing. I build networks, embeddings, and structured AI pipelines that connect evidence across scientific publications, technological designs, collaborations, and complex social and biological systems.

At Northwestern’s Center for Science of Science & Innovation, I study how ideas and technological capabilities develop: how fields change, how funding and collaboration shape discovery, and how research becomes useful technology. I combine large-scale data, network and language models, and domain knowledge. I also build open tools such as Helios Web to make these processes measurable, explorable, and useful.

Loading interactive network…
Helios Web visualization · Explore the tool at heliosweb.io

Current research

All projects
Graph and language-model representations used in an AI research pipeline.
Active research2026–present

Agentic AI for scientific and engineering design

Combining science-of-science evidence with structured extraction, network analysis, embedding search, and domain knowledge to identify promising scientific and engineering opportunities across publications, materials, measurements, and device concepts.

Agentic AIScience of scienceMaterialsDevices
Pipeline linking global science funding to research publications.
Active research2025–present

Mapping how science is funded

Linking funding acknowledgments, organizations, and publications at scale to study how public, philanthropic, and corporate support shapes scientific directions, researcher trajectories, and the geography of knowledge production.

Science fundingData infrastructureEntity resolution
Embedding map representing technical designs and performance signals.
Active research2025–present

Technology capability maps

Developing embedding models that represent design choices and technical attributes, reveal interpretable directions aligned with measurable progress, and help predict how technological capabilities evolve across materials and device platforms.

Technology forecastingEmbeddingsMaterialsDesign spaces
Helios Web network visualization mark.
Active open-source project2021–present

Helios Web

A browser-first platform for interactive exploration of large networks and embedding spaces, combining GPU rendering, a WebAssembly graph store, real-time layouts, and analysis across browsers, notebooks, and desktop workflows.

Network visualizationWebGPUWebAssemblyEmbeddings

Recent publications

Full record

Selected software

Software portfolio

Helios Web

Interactive network visualization at million-node scale

TP Similarity

Measure research-interest similarity with random walks