Senior AI Machine Learning Engineer at JERA Americas

Designed production reconciliation pipelines on Azure using Databricks, PySpark, ADF, Synapse, Delta Lake, SQL Server, and Oracle to replace manual row-level and aggregate validation with repeatable, auditable controls. Built anomaly-detection, root-cause, and exception-classification models using PyTorch, scikit-learn, gradient-boosting ensembles, and Azure Machine Learning to prioritize reconciliation breaks and explain likely failure causes. Translated control scenarios into parameterized validation rules, acceptance criteria, severity logic, and prioritized exception work queues consumed by business and support teams. Developed advanced T-SQL and PL/SQL stored procedures for reconciliation and dashboard serving, optimizing performance via partition switching, columnstore indexing, stats maintenance, and query tuning. Implemented cloud-native ingestion with ADF, Service Bus, and Functions, adding schema validation, idempotency, retries, and audit metadata. Enforced data quality with

Senior Machine Learning Engineer at Walmart Global Tech

Built automated reconciliation and validation pipelines with Azure Databricks, PySpark, ADF, Synapse, Delta Lake, SQL Server, and Oracle for high-volume product, order, inventory, shipment, pricing, and financial-control datasets. Developed scikit-learn and PyTorch anomaly detection and exception-classification models to detect unusual reconciliation patterns, assign probable root causes, rank business impact, and route prioritized work items. Created reusable rule engines for row counts, control totals, tolerance thresholds, duplicates, reference integrity, cross-system variances, and business-rule exceptions. Optimized T-SQL/PL-SQL workloads through indexing, partition-aware processing, stored-procedure refactoring, predicate/join tuning, and execution plan review. Implemented ingestion via ADF, Service Bus, and Azure Functions for APIs, databases, events, and file drops with validation, retries, dead-letter handling, lineage metadata, and controlled downstream triggers. Used Great E

Machine Learning Engineer at Assurex Health

Developed Azure-based reconciliation pipelines for claims, member, provider, laboratory, eligibility, and operational healthcare data using Databricks, PySpark, ADF, Delta Lake, SQL Server, Oracle, and Synapse. Built anomaly-detection and exception-classification models for duplicates, invalid identifiers, volume shifts, relationship breaks, and unexplained reconciliation variances. Created rule-based controls for row-level and aggregate validation, tolerance thresholds, reference checks, and date/code rules, producing prioritized exception queues for healthcare and audit-sensitive workflows. Authored T-SQL and PL/SQL procedures for staged loads and reconciliation with performance tuning for high-volume nightly processing. Implemented cloud-native ingestion with ADF, Service Bus, and Azure Functions including schema validation, audit metadata, asynchronous orchestration, and failure handling. Adopted Great Expectations and parameterized pytest suites for continuous validation and regre