Senior Data & Machine Learning Engineer

Job Description

Job Description

THE OPPORTUNITY

AKUVO is seeking a hands-on Senior Data & Machine Learning Engineer to build and own the production lifecycle of our proprietary predictive models and scores. This is a depth role: you are an exceptional model builder who can take a scoring problem from data through deployment largely single-handedly.

AKUVO’s model portfolio includes production and pilot capabilities supporting delinquency severity, propensity to pay, engagement, escalation, and a growing backlog of additional lending and collections use cases. You will build new models, enhance existing ones, and ensure they remain reliable, explainable, monitored, and ready for use within AKUVO IQ. You are a strong programmer who writes production-quality code, though this role focuses on model development rather than full-stack application or infrastructure engineering.

You will work closely with the Principal Data & Machine Learning Engineer, Data Engineering, Applied AI, financial-institution subject-matter experts, Product, and Compliance to translate business problems into defensible models that produce measurable value for AKUVO’s customers.

LOCATION

Local in Malvern/Philadelphia first, widening to surrounding areas such as New Jersey, New York, Delaware, while continuing to expand geographically in a hybrid/remote capacity based on location.

KEY RESPONSIBILITIES

• Own the design, development, validation, deployment, monitoring, and ongoing improvement of AKUVO’s predictive models and scores; build internal knowledge and ownership of existing production and pilot models through structured knowledge transfer, technical review, and documentation.
• Apply AKUVO’s four-phase Model Development Framework — Discovery & Design, Engineering R&D, Testing & Validation, and Deployment & Monitoring — across all score and attribute development work.
• Partner with business and financial-institution experts to define the problem, target outcome, prediction window, intended use, expected action, and measures of success for each model.
• Develop training datasets and features while addressing data quality, leakage, bias, missing values, class imbalance, and temporal consistency; design, train, compare, tune, and validate models appropriate for structured lending, portfolio, behavioral, and collections data.
• Evaluate model discrimination, calibration, stability, explainability, business value, and performance across relevant customer and portfolio segments.
• Establish reproducible experimentation, model versioning, model registry, approval, and release processes; build and maintain production pipelines for model training, scoring, deployment, rollback, monitoring, and retraining.
• Monitor model performance, drift, data changes, score distributions, stability, and operational outcomes; develop clear model documentation, technical specifications, model cards, assumptions, limitations, monitoring plans, and implementation guidance.
• Partner with the Data Engineering team on model-ready datasets, feature-source pipelines, lineage, and training-inference consistency; with the Principal Data & Machine Learning Engineer on technical guidance and review; with the Domain AI Analyst on business judgment, realistic scenarios, and acceptance criteria; with the Model Governance & Compliance Analyst on documentation, fair-lending review, and regulatory exam support; and with Product and Engineering to integrate model scores and attributes into AKUVO IQ.
• Use AI-assisted development tools and internal agents to accelerate research, feature exploration, coding, testing, documentation, and validation while maintaining appropriate technical review.

SKILLS AND EXPERIENCE

• 6+ years building, deploying, and supporting machine-learning models in production, with demonstrated end-to-end ownership of models developed largely single-handedly (problem definition → features → deployment → monitoring).
• Strong programming and production-quality coding in Python and SQL; comfortable developing, though not expected to own full-stack application or infrastructure engineering.
• Experience with machine-learning libraries such as scikit-learn, XGBoost, LightGBM, or comparable tools. Experience with PyTorch or TensorFlow is a plus.
• Strong experience with supervised-learning methods for classification, ranking, risk prediction, behavioral modeling, or similar structured-data problems.
• Experience with feature engineering, temporal validation, imbalanced datasets, model calibration, threshold selection, explainability, and performance analysis.
• Experience with Azure Machine Learning, Databricks, MLflow, or comparable cloud-based ML platforms; experience building reproducible training and inference pipelines, model registries, automated tests, CI/CD, and production monitoring.
• Strong understanding of model drift, data drift, stability, performance degradation, retraining, and production troubleshooting.
• Ability to translate business objectives into clearly defined modeling problems, and to communicate model methodology, performance, limitations, and intended use to technical and nontechnical audiences.
• Sound software-engineering practices (source control, testing, documentation, modular design), cross-functional collaboration, and active use of AI-assisted tools to improve productivity and quality.

PREFERRED QUALIFICATIONS

• Experience developing credit-risk, lending, collections, delinquency, propensity, engagement, loss, or financial-behavior models.
• Experience working with credit unions, banks, fintech, servicing, or other regulated financial-services organizations.
• Experience with model governance, independent validation, fair-lending analysis, adverse-action considerations, or regulatory model-risk expectations.
• Experience with explainability techniques, bias and fairness testing, challenger models, champion-challenger frameworks, or model stress testing.
• Experience with feature stores, distributed processing, containers, workflow orchestration, or ML-observability platforms.
• Experience with Microsoft Fabric, OneLake, Azure Synapse, Azure DevOps, or the broader Microsoft data ecosystem.
• Experience integrating model outputs into B2B SaaS products, APIs, decisioning systems, or operational workflows.
• Bachelor’s or advanced degree in computer science, statistics, mathematics, data science, engineering, economics, or a related quantitative field, or equivalent practical experience.