Kajal Choudhary

HSBC credit risk project: designed and built end-to-end ETL pipelines in Databricks/PySpark and dbt on GCP; built RAG solutions (embeddings, vector search, LLM APIs) and AI agent workflows with tool calling and MCP to automate data validation, reconciliation and documentation; risk data marts, dimensional modelling, Power BI/Tableau reporting datasets.

• Converted business requirements into comprehensive technical specifications for data models, KPI catalogues and reporting pipelines, collaborating with engineering and BI teams to ensure successful implementation. • Execute SQL-based data analysis and validation across retail banking datasets using Oracle and SQL Server, performing data profiling, quality assessments, completeness checks and consistency analysis to identify and resolve data issues.

Conducted large-scale SQL querying and advanced data analysis across 100TB+ Oracle and SQL Server financial datasets, extracting, validating and analysing complex data to identify credit risk trends, data quality issues and actionable insights for downstream teams. • Designed and maintained comprehensive source-to-target mapping documentation and transformation specifications for ETL migration and development initiatives, including field-level mappings, data type conversions, business rules and validation

• Analysed large-scale e-commerce and digital retail datasets using Google Analytics 4, Google Ads and Adobe Analytics, extracting, validating and documenting insights to support strategic decision-making, customer behaviour analysis and campaign performance optimisation.