Uploaded April 2026 | Updated September 2026, 2 weeks ago
Financial systems demand strong security guarantees while still enabling high-value analytics. This session examines how homomorphic encryption (HE) and GAN-based synthetic data can advance secure financial intelligence in alignment with OWASP’s mission of building trustworthy systems.
Homomorphic encryption enables computation on encrypted data without ever decrypting it, eliminating exposure of sensitive financial records during analysis. Modern GPU-optimised HE frameworks now support SQL-like queries and lightweight machine-learning inference directly on encrypted transaction datasets, bringing performance close to traditional execution and reducing operational overhead previously associated with secure multi-party approaches. Integrating HE into ETL and analytics workflows strengthens defences against data exfiltration, insider threats, and misconfigured infrastructure, while supporting GDPR and PCI-DSS compliance.
Complementing HE, GAN-powered synthetic data generation using models such as TimeGAN and Tabular GAN creates statistically realistic financial datasets that preserve behavioural patterns, including rare anomalies and seasonal variation, without retaining identifiable information. These synthetic datasets allow teams to develop fraud detection models that converge faster, conduct secure cross-team data sharing, and perform stress testing without risking real customer data.
Leela krishna Yenigalla
Senior Software and Data Engineer
Leela Krishna Yenigalla is a seasoned Senior Software and Data Engineer with over 10 years of experience in designing and implementing scalable ETL pipelines and data solutions across cloud and on-premises environments. He specialises in PySpark, Snowflake, and AWS, with a strong track record of integrating structured and semi-structured data for high-performance analytics and reporting. Leela has expertise in ingesting and transforming large datasets from diverse sources such as S3, APIs, relational databases, and Hadoop into Snowflake and cloud data lakes. He has led large-scale, data-intensive enterprise BI initiatives integrating multi-terabyte backends, harmonising disparate sources, and delivering production-grade, high-performance BI applications. His expertise spans data modelling, query optimisation, data quality assurance, and automation using tools like Apache Airflow, AWS Glue, and Step Functions. He also brings a wealth of experience in BI development with Tableau, Power BI, and Web FOCUS. Throughout his career, Leela has worked with leading organisations, including Worldpay and Ozburn Hessey Logistics, where he contributed to data modernisation efforts and performance tuning of data pipelines. He is proficient in a wide array of technologies, including SQL, Python, Java, Informatica, SSIS, Docker, Kubernetes, and cloud platforms like AWS, GCP, and Azure.
-
Managed by the OWASP® Foundation
owasp.org
Financial systems demand strong security guarantees while still enabling high-value analytics. This session examines how homomorphic encryption (HE) and GAN-based synthetic data can advance secure financial intelligence in alignment with OWASP’s mission of building trustworthy systems.
Homomorphic encryption enables computation on encrypted data without ever decrypting it, eliminating exposure of sensitive financial records during analysis. Modern GPU-optimised HE frameworks now support SQL-like queries and lightweight machine-learning inference directly on encrypted transaction datasets, bringing performance close to traditional execution and reducing operational overhead previously associated with secure multi-party approaches. Integrating HE into ETL and analytics workflows strengthens defences against data exfiltration, insider threats, and misconfigured infrastructure, while supporting GDPR and PCI-DSS compliance.
Complementing HE, GAN-powered synthetic data generation using models such as TimeGAN and Tabular GAN creates statistically realistic financial datasets that preserve behavioural patterns, including rare anomalies and seasonal variation, without retaining identifiable information. These synthetic datasets allow teams to develop fraud detection models that converge faster, conduct secure cross-team data sharing, and perform stress testing without risking real customer data.
Leela krishna Yenigalla
Senior Software and Data Engineer
Leela Krishna Yenigalla is a seasoned Senior Software and Data Engineer with over 10 years of experience in designing and implementing scalable ETL pipelines and data solutions across cloud and on-premises environments. He specialises in PySpark, Snowflake, and AWS, with a strong track record of integrating structured and semi-structured data for high-performance analytics and reporting. Leela has expertise in ingesting and transforming large datasets from diverse sources such as S3, APIs, relational databases, and Hadoop into Snowflake and cloud data lakes. He has led large-scale, data-intensive enterprise BI initiatives integrating multi-terabyte backends, harmonising disparate sources, and delivering production-grade, high-performance BI applications. His expertise spans data modelling, query optimisation, data quality assurance, and automation using tools like Apache Airflow, AWS Glue, and Step Functions. He also brings a wealth of experience in BI development with Tableau, Power BI, and Web FOCUS. Throughout his career, Leela has worked with leading organisations, including Worldpay and Ozburn Hessey Logistics, where he contributed to data modernisation efforts and performance tuning of data pipelines. He is proficient in a wide array of technologies, including SQL, Python, Java, Informatica, SSIS, Docker, Kubernetes, and cloud platforms like AWS, GCP, and Azure.
-
Managed by the OWASP® Foundation
owasp.org
