Orchestrated enterprise ETL/ELT pipelines using PySpark on Databricks, AWS Glue, Matillion, and Snowflake across procurement and finance domains, processing 20+ source tables and supporting 9 downstream analytics datasets.

Optimized cloud data transformations using PySpark, SQL, and Python within Snowflake and Databricks environments across 14 production workflows, supporting ML- and AI-ready analytics and reducing execution latency across 6 reporting cycles.

Implemented secure data integrations using REST APIs, OAuth, and JWT across internal and third-party systems, enabling LLM-assisted analytics and AI-driven data consumption.

Constructed analytics-ready data models using Snowflake views and relational schemas to support enterprise reporting and RAG-based analytical retrieval, enabling 11 standardized datasets consumed by 4 BI platforms.

Monitored data pipeline reliability through validation checks, logging, and error handling across 16 production batch workflows, supporting agentic AI-assisted exception handling and preventing 12 recurring data quality issues.

Supported applied machine learning and AI use cases by preparing feature-ready datasets through PySpark transformations on Databricks, enabling downstream risk scoring and pattern detection workflows.

Coordinated with business analysts, system owners, and IT teams across 7 cross-functional initiatives, aligning analytics, AI/ML, and data requirements for 18 documented use cases.

Analyzed procurement, inventory, and supplier risk data using SQL, Python, and MongoDB, reviewing 10 operational tables to support 4 recurring risk assessment and reporting workflows.

Applied machine learning techniques using Python to historical supplier and transaction data, executing 2 risk analysis use cases

that surfaced 15 risk indicators for business review and prioritization.

Prepared risk-focused analytical datasets through data validation, normalization, and aggregation across procurement systems, maintaining 6 compliance-ready datasets used in compliance and risk reporting.

Automated recurring risk and compliance reports using SQL scripts and Python within scheduled daily jobs, maintaining 12 operational reports and eliminating 6 manual reconciliation steps.

Collaborated with risk analysts, procurement teams, and IT partners during data review and reporting cycles, engaging 5 cross-functional teams across 9 documented risk review sessions.

Built machine learning data pipelines using Python, AWS S3, Redshift, and Databricks within cloud analytics environments, across 6 structured data sources, enabling 4 AI-driven analytical use cases.

Trained machine learning models using scikit-learn on curated healthcare and financial datasets, across 5 classification and anomaly detection experiments, producing 12 validated model outputs.

Prepared feature-ready datasets through feature engineering, scaling, and encoding across regulated data domains, across 7 modeling iterations, removing 90 manually data preparation hours.

Integrated AI model outputs into analytics workflows using Python and SQL server within batch processing jobs, across 3 reporting pipelines, enabling 6 downstream analytical reports.

Collaborated with data engineers, analysts, and domain stakeholders during model development and validation cycles, across 4 cross-functional teams, improving delivery alignment across 10 project milestones.

 Assisted graduate admission team with the admission process and other academic requirements.

 Maintained admissions reporting workflows using SQL, Python, and Power BI within enrollment and applicant tracking systems, across 8 historical admissions datasets, reducing application processing by 5 admission cycles.

 Applied machine learning classification techniques using Python on applicant, academic, and demographic data, across 3 admissions analysis use cases, identifying 900 high-probability applicants.

Engineered analysis-ready datasets from multi-year admissions records through cleaning and normalization, supporting 5 modeling initiatives and removing 140 manual preparation hours.

Streamlined enrollment tracking workflows by automating reporting logic with SQL and Python, improving operational visibility across 18 recurring reports and removing 9 manual reconciliation steps.

Conducted exploratory data analysis on applicant and enrollment datasets using trend and distribution analysis, reviewing 12 datasets and identifying 16 data quality issues.

Developed data analysis workflows using SSIS, Python, Tableau, and Power BI within centralized reporting environments, across 12 datasets for 5 teams, reducing report preparation by 10 reporting cycles.

Applied machine learning methods including linear regression and clustering using Python on historical business data, across 6 forecasting and segmentation tasks, identifying 18 recurring data patterns.

Built analysis-ready datasets from transactional systems through transformation and normalization, enabling 8 modeling efforts and eliminating 240 manual data preparation hours.

Created scalable batch reporting and data refresh pipelines using SQL and Python, supporting 30 recurring reports and eliminating 14 manual data pulls across business teams.

Designed dimensional data models using fact and dimension tables to standardize business metrics, supporting consistent reporting for 8 stakeholder groups across 4 tools.

Performed exploratory analysis across structured enterprise datasets using distribution profiling and trend analysis, examining 18 datasets and flagging 22 inconsistencies.

Supported data validation processes by implementing row count checks and reconciliation logic across ingestion layers, monitoring 10 data feeds, preventing 9 reporting interruptions.