Metadata Analyst (Scribe) – Machine Learning focus

About the Company

IBM is a global technology and consulting company headquartered in Armonk, New York, with a major presence across the globe. Known for its extensive portfolio of hardware, software, and services, IBM is a leader in areas such as artificial intelligence, cloud computing, and quantum computing. We are committed to fostering innovation that matters for our company and for the world. Join a team where you can make an impact.

Job Description

We are seeking a highly detail-oriented and analytical Metadata Analyst with a strong focus on Machine Learning to join our innovative team. In this pivotal ‘Scribe’ role, you will be responsible for defining, capturing, and managing metadata associated with complex machine learning models, datasets, and experiments. Your work will ensure the discoverability, interpretability, and governance of our ML assets, playing a critical part in the development and deployment of robust and ethical AI solutions. This role requires a deep understanding of data structures, ML lifecycles, and a passion for organizing information to drive AI excellence.

Key Responsibilities

• Develop and implement metadata standards and taxonomies specifically for machine learning models, datasets, features, and experiment configurations.
• Act as the primary 'Scribe' for ML projects, systematically documenting metadata from ideation through deployment and monitoring.
• Collaborate closely with ML Engineers, Data Scientists, and Researchers to capture essential metadata during the model development lifecycle.
• Ensure data lineage, model versioning, and explainability metadata are accurately recorded and easily accessible.
• Design and maintain metadata repositories and tools to support efficient search, retrieval, and analysis of ML assets.
• Conduct metadata quality checks and implement governance procedures to maintain accuracy and completeness.
• Educate and train ML teams on best practices for metadata capture and management.
• Contribute to the continuous improvement of ML MLOps pipelines by integrating metadata management solutions.

Required Skills

• 3+ years of experience in data analysis, metadata management, or information architecture.
• Proven understanding of machine learning concepts, models, and data pipelines.
• Familiarity with data governance principles and practices.
• Proficiency in SQL and experience with relational or NoSQL databases.
• Strong analytical skills with an ability to interpret complex data structures.
• Excellent communication and collaboration skills.
• Bachelor's degree in Computer Science, Data Science, Information Systems, or a related quantitative field.

Preferred Qualifications

• Master's degree in a relevant field.
• Experience with MLOps platforms (e.g., MLflow, Kubeflow) or data cataloging tools.
• Knowledge of schema-on-read/write principles and data modeling for semi-structured data.
• Familiarity with programming languages like Python for data manipulation and scripting.
• Experience with cloud platforms (AWS, Azure, GCP) and their data/ML services.

Perks & Benefits

• Comprehensive health, dental, and vision insurance.
• Generous paid time off and holidays.
• 401(k) retirement plan with company match.
• Professional development opportunities and tuition reimbursement.
• Access to cutting-edge AI and ML research facilities.
• Flexible hybrid work arrangements.
• Employee assistance programs.
• Wellness initiatives and on-site amenities (where applicable).

Apply Now