Selected work
- Python
- Flask
- JavaScript
- Elasticsearch
- Open source
- In production
- ~30k lines of code
A research platform for the ERC Starting Grant project
VERITRACE (University of Lisbon). It gives
historians search and analysis across more than 413,000 historical texts in
six languages, running on cloud infrastructure.
- Data pipelines that harmonise metadata from different sources into a common Dublin Core schema.
- Sentence-level vector embeddings that find semantically similar passages across texts.
- Advanced full-text search over the whole corpus.
Wolf Labs is the project's external technical partner and designs, builds and
maintains its technical infrastructure, data pipelines, analytics and research platform.
Independent research projects
Self-initiated academic projects, not client work. They show the kind of tools
I build when the research question is my own.
Personal research project
- Concordance
- Semantic search
- Text matching
- MIT-licensed software
- Work in progress
Tools for studying David Hume's writings across his full textual corpus:
search, compare and map his texts, with results that cite back to the
open-access edition on davidhume.org.
Personal research project
- Concordance
- Keyness analysis
- Custom subcorpora
- Public release v1.1
A companion tool for studying the clinical correspondence of Dr William
Cullen: about 5,600 eighteenth-century consultation letters, digitised by the
AHRC-funded Cullen Project at Glasgow. It offers concordance searches,
keyness comparisons between subcorpora, and curated analyses of voice,
disease, geography and Cullen's consulting network.
Services
-
Research platforms
Design, development and maintenance of full-stack web platforms for scholarly projects.
-
Data pipelines and infrastructure
Ingestion, cleaning, enrichment and hosting of large historical and bibliographic datasets.
-
Search, analytics and machine learning
Full-text and semantic search, and Python-based analysis of text and metadata for historical research.
-
Knowledge graphs and data presentation
Reviewing data models and recommending how to present data of uneven quality to researchers.
-
Documentation and handover
Instructions and training so that teams can continue the work in-house.
How I work
Scope and deliverables are agreed up front. I deliver against defined milestones and document the results so that your
team can maintain them without me.
About
Jeffrey Wolf holds a PhD in the history of medicine from the University of
Edinburgh. Before that he studied philosophy at Princeton University and the
philosophy and history of science at the University of California, Berkeley.
He has taught web development to business school students at Northwestern
University, and works in Python, data analytics, machine learning and
full-stack web development.
Contact
jeffwolf@wolf-labs.eu
Based in Berlin.
