Job Title

We are seeking a hands-on Data Scientist to develop a matching and recommendation capability that identifies the most likely application associated with a DNS record. The candidate will combine DNS, CMDB, application, server, IP, ownership, and existing mapped-record reference data, then apply appropriate data science techniques to generate ranked application matches with explainable confidence scores. The solution should reduce manual investigation and support integration into the broader remediation workflow.

This role supports the Orphan DNS Remediation initiative by applying data science to associate DNS records with the correct enterprise applications.

Key Responsibilities

• Work with ISRM (Information Security Risk Management) stakeholders and domain specialists to define the DNS-to-application matching problem, business rules, and measurable success criteria.
• Profile, cleanse, normalize, join, and validate data from DNS, CMDB, application, server, IP, ownership, and mapped-record reference sources.
• Design and compare appropriate matching approaches, including deterministic rules, fuzzy or similarity matching, entity resolution, classification, clustering, and graph-based analysis where relevant.
• Develop a recommendation engine that returns ranked candidate applications with confidence scores and supporting evidence.
• Validate the solution using confirmed historical mappings and metrics such as precision, recall, top-k accuracy, coverage, and false-match rate.
• Perform error analysis and improve features, algorithms, thresholds, and data-quality rules based on validation results and stakeholder feedback.
• Build reusable Python notebooks and reproducible model pipelines using approved Azure and/or AWS services, with clear technical documentation and knowledge transfer.