AVP, Applied Model Ops Developer

Role Overview

• Engage regularly with model developers, validators, and risk stakeholders to understand their evolving data needs for model development, monitoring, and governance
• Partner with credit analytics, risk, fraud, marketing, and operations functions to identify, define, and prioritize use cases requiring model-ready data
• Build scalable data architectures to support real-time and batch monitoring, including data ingestion, enrichment, and retention practices
• Support pipeline development by designing and maintaining automated end-to-end ML pipelines for data collection, preprocessing, feature engineering, and model training
• Conduct data transformation by converting raw observations into variables (features) that machine learning models can understand, such as turning timestamps into cyclical time features
• Transforming theoretical data science prototypes into robust, high-performance software systems that can handle large volumes of real-time data
• Build and maintain automated pipelines that handle not just code, but also data validation, model training, and artifact management
• Design, develop, and maintain robust pipelines to collect, transform, and store data used in model monitoring workflows (e.g., scoring data, performance metrics, outcomes)
• Provide thought and technical leadership in generating new signals from raw data by applying techniques such as normalization, scaling and categorical encoding
• Integrate data pipelines with model lifecycle platforms, MLOps tools, and observability solutions to ensure seamless model performance tracking
• Partner with model risk and compliance teams to ensure data lineage, audit trails, and documentation are preserved and accessible for regulatory reviews (e.g., SR 11-7 compliance)
• Liaise with cloud, data lake, data warehouse, and model governance engineering teams on delivery execution and backlog prioritization
• Collaborate with data scientists, model validators, and product managers to align monitoring data infrastructure with evolving model monitoring requirements
• Optimize data storage and compute performance for large-scale monitoring use cases involving high-frequency scoring or model ensembles

Requirements

• Bachelor’s degree in a quantitative, technical, or data-focused field (e.g., Statistics, Mathematics, Computer Science, Data Science, Engineering) with 6+ years’ experience OR in lieu of a degree 8+years of relevant work experience in monitoring, validation, or credit risk strategy
• Minimum 6+ years of professional experience in model operations, data engineering, or analytics infrastructure
• Strong proficiency with data engineering tools and frameworks (e.g., Apache Spark, Airflow, Kafka, dbt, PySpark)
• Proficient in programming languages such as SAS, Python, and SQL for building monitoring pipelines and validation checks
• Experience with cloud-based data infrastructure (e.g., AWS, Azure, GCP) and data warehousing (e.g., Snowflake, Redshift, BigQuery)
• Familiarity with MLOps practices, model metadata tracking (e.g., MLflow), and monitoring toolkits (e.g., Evidently AI, WhyLabs, Prometheus)
• Understanding of model risk governance requirements and the role of data engineering in ensuring compliant model monitoring
• Ability to work in an agile environment and deliver high-quality, production-grade code in collaboration with DevOps and platform engineering teams

Tech Stack

• Airflow
• Amazon Redshift
• Apache
• AWS
• Azure
• BigQuery
• Cloud
• Google Cloud Platform
• Kafka
• Prometheus
• PySpark
• Python
• Spark
• SQL

Benefits

• best-in-class employee benefits
• programs that cater to work-life integration and overall well-being
• career advancement and upskilling opportunities