• Engineered data pipelines and quality checks to manage 22 freight carriers amounting to $20M per month of invoices using Django/Python backend, ClickHouse columnar database, and Dagster scheduler on AWS• Saved close to $75,000 monthly for another SMB client by launching multi-invoice download functionality using React/TypeScript and Django/Python• Optimized and integrated Amazon Textract and OpenAI’s LLM to parse invoices for accessorial charges, pioneering its first production use and enhancing automated invoice auditing

Part of the Data Analytics/Data Warehousing team, designed and migrated 7 Human Capital datasets into a real-time data warehouse using Java, Spring Boot, Spring Batch, and MS SQL Server. Developed and launched a BI tool for the executive team, which expanded to 20+ Directors within 2 months of launchKey achievements:1. Enhanced the Human Capital Data Warehouse (HC DWH) with 2 sensitive and 2 app usage datasets through ETL pipelines developed using Java, Spring Batch/MVC/Boot, and a Kimball model-based schema in SQL Server2. Developed an HC GraphQL Java API to deliver 10 critical insights on hiring events, with two aggregated and detailed data nodes, reducing computational burdens for HC data analysts3. Transformed 5 distinct Excel-based attrition reports spanning over 5 years of HC data into a centralized and real-time analytics dashboard using React and TypeScript, eliminating months of manual report generation and upkeep for the Executive leadership team4. Spearheaded the collaboration with 4 stakeholder teams in an Agile environment for the successful launch of an HC intelligence tool delivering key hire, headcount, and attrition trends across 20+ metrics, facilitating its expansion toover 10 firm-wide Directors within 2 months of its introduction5. Optimized the tool’s computations by 50% leveraging React memo on the frontend and EH Cache on the backend6. Established a robust monitoring infrastructure for the DWH’s 7 services, leveraging Python, CRON jobs, and Jenkins CI/CD pipeline to ensure seamless deployment and swift downtime resolutionTech stack: 1. Backend: Java, SQL Server, MyBatis, Spring framework (MVC, Batch, Boot), Drools engine, GraphQL2. Frontend: JavaScript, TypeScript, React, Redux, Highcharts, Pivots3. System: Linux, Git, Kerberos, Load balancers, Prometheus metrics, Monitoring setup

As part of the Data Analytics/Data Warehouse team, developed a hybrid tool in React/Redux to fetch human capital (HC) data from GraphQL APIs and perform complex analyses using Pivots and Charts. Undertook additional task of improving GraphQL (Java) backend performance via Caching, and GraphQL query processing to accurately generate and process nested sorting via MyBatis.