AI-Native Metadata Librarian

Third Iron•Minneapolis, MN

5d•Remote

About The Position

Do you enjoy digging into complex bibliographic data, tracking down inconsistencies, and finding better ways to organize information at scale? Are you the person your colleagues turn to when metadata does not behave as expected—or when a large dataset needs to be understood, cleaned, or transformed? Do you enjoy using Excel, SQL, and AI tools to answer questions and develop more efficient workflows? If so, we would love to talk with you about joining Third Iron as an AI-Native Metadata Librarian. Since our remote-first company's founding in 2011, Third Iron has created the industry-leading software services BrowZine and LibKey. Our users include millions of students, faculty, doctors, researchers, and scientists affiliated with more than 2,000 libraries at universities, hospitals, corporations, and government departments across 36 countries—including organizations such as Stanford University, Mayo Clinic, NHS England, and the University of Hong Kong. In this role, you will help improve and evolve the bibliographic data that powers Third Iron's systems and services. Working at the intersection of librarianship, cataloging, data analysis, and emerging technology, you will explore large-scale bibliographic databases, investigate complex metadata issues, and translate your findings into practical improvements for our products, processes, and users. You will use Excel, SQL, and AI-assisted tools while collaborating with technical and product teams to strengthen metadata quality, coverage, matching, and interoperability.

Requirements

At least 2 years of professional work experience in libraries, data science, metadata analysis or similar role.
Master's degree in library and information science, a related field, or equivalent professional experience.
Professional experience working with bibliographic metadata, library systems, discovery services, knowledge bases, or scholarly information.
Strong working knowledge of Excel, including functions, filters, lookups, data transformation, and analysis of large datasets.
High degree of comfort using AI tools for research, analysis, vibe coding mini apps, workflow development, and problem-solving.
Familiarity with bibliographic identifiers and metadata standards such as DOI, ISSN, ISBN, MARC, KBART, Dublin Core, or related formats.
Strong analytical skills and the persistence to investigate ambiguous or incomplete data.
Ability to explain metadata and data-quality issues clearly to both technical and nontechnical colleagues.
Excellent written communication, documentation, and remote-collaboration skills.
Experience with APIs, JSON, XML, regular expressions, scripting, or data-transformation tools.
Comfortable with reviewing metadata samples and designing tools to analyze data mechanically with AI.
Ability to work independently, manage priorities, and contribute effectively within a distributed team.

Nice To Haves

Experience writing SQL queries to explore, join, validate, and analyze data.
Familiarity with metadata supplied by publishers, content aggregators, link resolvers, or library vendors.
Knowledge of entity resolution, record matching, deduplication, authority control, or identifier reconciliation.
Familiarity with scholarly communication infrastructure and data sources such as Crossref, PubMed, OpenAlex, ORCID, or library knowledge bases.
Experience evaluating or developing AI-assisted metadata workflows.
Experience collaborating with software engineers, product managers, or data teams.
1+ year of success working in a remote position.

Responsibilities

Analyze, clean, reconcile, and enrich large-scale bibliographic and holdings datasets.
Investigate metadata quality, duplication, normalization, matching, and entity-resolution issues.
Use Excel, SQL, AI tools, and other data-analysis methods to identify patterns, exceptions, and opportunities for improvement.
Help design and refine metadata rules, mappings, validation processes, and automated workflows.
Evaluate bibliographic data from publishers, aggregators, libraries, and other external sources.
Support the evolution of Third Iron's metadata architecture, discovery capabilities, and knowledge-base systems.
Develop repeatable methods for assessing metadata completeness, accuracy, consistency, and timeliness.
Document data sources, standards, workflows, decisions, and quality-control procedures.
Use AI-assisted approaches thoughtfully while validating results and maintaining appropriate standards for accuracy, privacy, and responsible use.