Library Scientist

Miratech helps visionaries change the world. We are a global IT services and consulting company that brings together enterprise and start-up innovation. Today, we support digital transformation for some of the world's largest enterprises. By partnering with both large and small players, we stay at the leading edge of technology, remain nimble even as a global leader, and create technology that helps our clients further enhance their business. We are a values-driven organization, and our culture of Relentless Performance has enabled over 99% of Miratech's engagements to succeed by meeting or exceeding our scope, schedule, and/or budget objectives since our inception in 1989.

Miratech has coverage across 5 continents and operates in over 25 countries around the world. Miratech retains nearly 1000 full-time professionals, and our annual growth rate exceeds 25%.

The Library Scientist builds the information architecture that Genesys Cloud AI runs on. The client has been explicit that taxonomy, searchability, and content structure are the critical prerequisites for AI knowledge surfacing in Genesys Cloud and for any later bot development. No amount of platform capability compensates for a knowledge base that is unstructured, inconsistently tagged, or unsearchable - the AI will retrieve from it either way, and the quality of what it returns is set here.

This is a specialist, deliverable-heavy role, and the most technically consequential of the three for AI outcomes. The successful candidate is a trained information professional who can analyse a real, messy, high-volume content set and produce a taxonomy and metadata model that holds up under production load and under machine retrieval.

Responsibilities:

• Analyse the current knowledge base content and structure. Assess the existing content inventory - volume, format, ownership, age, duplication, structural consistency, and current tagging practice. Document the current state and the specific structural problems that are suppressing findability.
• Conduct a Baseline Content Analysis of 100-200 pilot articles. Perform a detailed article-level assessment across the pilot set: accuracy, currency, readability, reading level, completeness, duplication, structural consistency, and tagging quality. Produce scored findings and a remediation profile that can be extrapolated to the full corpus.
• Execute a Search Behaviour Analysis. Analyse actual search logs and agent and self-service search behaviour - query volume, top queries, zero-result and low-result queries, abandonment, refinement patterns, and the vocabulary gap between how users search and how content is written. Translate findings into concrete taxonomy and synonym requirements.
• Build the Taxonomy Blueprint, designed for AI retrieval. Design the taxonomy - hierarchy, facets, controlled vocabularies, synonym rings, and the rules governing where content sits and how it relates. Design for the dual audience of internal agents and public self-service, and for the third consumer that will matter most in Phase 4: the Genesys AI layer that retrieves, summarises and answers from this corpus. Validate the structure with business line subject matter experts and test it against real queries.
• Structure content for machine consumption. Specify how articles should be scoped, chunked and written so that Genesys AI knowledge surfacing and virtual agents retrieve the right passage rather than the right-ish document. Define article granularity, heading structure, answer-first writing patterns, and the disambiguation rules that prevent two near-identical articles from producing contradictory AI answers.
• Define the metadata and tagging model. Specify the metadata schema - required and optional fields, value sets, inheritance rules, audience and jurisdiction markers, lifecycle and review-date fields, and the tagging conventions authors will follow. Define how metadata drives search relevance, filtering, and content targeting.
• Optimise searchability across both keyword and semantic retrieval. Specify and test the search configuration - relevance tuning, synonym and acronym handling, stemming, boosting, and result presentation - and assess how the content performs under Genesys Cloud's natural-language and semantic search as distinct from literal keyword matching. Establish baseline search performance measures and the target improvement, and define the measures by which AI answer quality will be judged once knowledge surfacing is enabled.
• Establish content governance at the information level. Define content standards, lifecycle and review cadence, ownership at the article and category level, and the quality criteria content must meet to be published. Work with the Senior Business Consultant so that the Governance Playbook and the taxonomy are a single coherent model rather than two parallel documents.
• Observe proofs of concept with a specialist eye. Attend PoC sessions and test, specifically, whether Genesys Cloud can carry the complex tagging structures, faceted search, and multi-audience content model that will be built in Phase 4 - and whether its AI retrieval returns accurate, attributable answers from the pilot corpus. This is the single most important technical validation in the evaluation, and it is this role's responsibility to make the call.
• Transfer knowledge to the client's content team. Document the model so that the client's own authors and content owners can apply it, and run the working sessions that build that capability.

• Master of Library and Information Science (MLIS/MISt) or an equivalent information science qualification. This is a genuine requirement, not a preference - the role is named for the discipline.
• 5+ years applying information science in an enterprise or digital context: taxonomy design, metadata modelling, controlled vocabularies, information architecture, or enterprise search.
• Demonstrated experience designing and implementing a taxonomy and metadata model for a production content environment, with examples of the deliverables produced.
• Practical experience with enterprise search behaviour analysis - working with search logs, diagnosing zero-result and relevance failure, and tuning search configuration.
• Working knowledge of taxonomy and metadata standards and the ability to apply them pragmatically (SKOS, Dublin Core, ISO 25964, schema.org).
• Demonstrable understanding of how information architecture affects AI and natural-language retrieval quality - why structure, granularity, duplication and tagging determine what a knowledge-surfacing or virtual agent capability returns.
• Ability to conduct structured, scored content analysis at article level and report findings clearly to a non-specialist audience.
• Eligible to obtain and hold a Government of Canada security clearance.

Nice to have:

• Hands-on experience with the Genesys Cloud knowledge base and its authoring and optimisation tooling. Failing that, a comparable contact centre knowledge platform - Salesforce Knowledge, ServiceNow Knowledge Management, Zendesk Guide.
• Direct experience preparing a content corpus for AI, retrieval-augmented generation, or virtual agent consumption - chunking, structure, grounding, and the relationship between content hygiene and model output quality.
• Experience establishing AI answer-quality evaluation - measuring accuracy, attribution and failure modes of a retrieval-based assistant against a known content set.
• Government, public sector, or academic library background, with familiarity with plain-language standards, WCAG accessibility requirements, and records management obligations.
• Bilingual English/French, including experience managing a bilingual taxonomy and parallel content sets.
• Experience with KCS (Knowledge-Centred Service) methodology.

Additional information

We offer:

• Culture of Relentless Performance: join an unstoppable technology development team with a 99% project success rate and more than 30% year-over-year revenue growth.
• Competitive Pay and Benefits: enjoy a comprehensive compensation and benefits package, including health insurance, language courses, and a relocation program.
• Growth Mindset: reap the benefits of a range of professional development opportunities, including certification programs, mentorship and talent investment programs, internal mobility and internship opportunities.
• Global Impact: collaborate on impactful projects for top global clients and shape the future of industries.
• Welcoming Multicultural Environment: be a part of a dynamic, global team and thrive in an inclusive and supportive work environment with open communication and regular team-building company social events.
• Social Sustainability Values: join our sustainable business practices focused on five pillars, including IT education, community empowerment, fair operating practices, environmental sustainability, and gender equality.

* Miratech is an equal opportunity employer and does not discriminate against any employee or applicant for employment on the basis of race, color, religion, sex, national origin, age, disability, veteran status, sexual orientation, gender identity, or any other protected status under applicable law.