HEMANT VERMA

Software Engineer - III

American Express•  July 2022 - Present

• Engineered a highly-effective solution using Postgres Triggers to prevent duplicate communications to customers, improving their overall experience. • Designed an ETL-like framework using S3, Apache Kafka, and Postgres to process batch use cases such as marketing campaigns and account snapshots from various sources like SFTP, REST API etc. • Collaborated in design and architectural reviews to influence solutions for complex design problems. • Led discussions involving cross-border teams and POs to align on deliverable timelines.

Software Engineer - II

American Express•  December 2020 - June 2022

• Designed and implemented horizontally scalable distributed micro-services for Enterprise Communication Platform that delivers over 30 million compliance-enabled communications daily, peaking at 1500 TPS. • Designed and implemented Priority processing and Throttling using Eclipse Vert.x framework and RxJava, with the capability to handle various SLA buckets and reprocessing of the failed communication events. • Devised a zero-deployment, configuration-based solution for new communication use case on-boarding, reducing development efforts and marketing time by 90%. • Designed and implemented Postgres Partitioning achieving 700 times improvement in Poller query performance. • Implemented a new system to comply with European regulations and send Forex rates to customers within a sprint, saving €5 million in potential risk. • Mentored interns and led the design in developing a mock service from scratch that mocks various dispatcher responses, ultimately saving costs for millions of SMS/emails during performance/load testing. • Designed and implemented automated stale events detection.

Software Engineer - I

American Express•  July 2019 - November 2020

• Implemented a highly scalable and reliable Event-Driven Architecture using Apache Kafka message broker. • Designed and implemented deployment strategy for Kafka on cloud-platform and graceful shutdown for critical Tier 1 applications. • Collaborated with cross-functional teams to quickly resolve production issues, minimizing customer impact and avoiding potential CAPs and OREs. • Performed base-lining and performance testing for multiple services.

Machine Learning Intern

Neva Innovation Labs•  May 2018 - July 2018

• Developed a Driver Behaviour profiling application to detect various events generated by car driver such as braking, acceleration using mobile sensor data. • Implemented a random forest model with a sliding window approach to extract feature vectors. • Achieved 93% accuracy by further developing an LSTM classification model to detect different types of events.