Farhath Banu

About

Backend/data engineer who builds production FastAPI microservices and AWS serverless/Glue pipelines for SMS analytics and marketing segmentation. Led a legacy batch modernization into modular services (FastAPI + Glue/Athena + ClickHouse) using shadow-mode parity checks, feature flags, and incremental rollout. Demonstrated measurable performance wins (12s to sub-second SQL; ~40% CPU reduction) and strong incident ownership with proactive schema-drift prevention.

Experience

Education

Key Strengths

• Built and maintained production FastAPI microservice for real-time SMS analytics (Kafka/SQS ingestion, validation/enrichment, persistence)
• Designed resilient distributed processing with Celery retries and VPC-isolated internal workers
• Implemented unified authentication layer across multiple providers (Google SSO, email/password, SMS auth)
• Strong testing discipline: TDD, high code coverage, Cypress integration testing, and load testing
• Delivered AWS serverless ingestion/processing using Lambda + SQS with Serverless Framework and GitHub Actions CI/CD
• Built production AWS Glue ETL (JSON from S3 to partitioned Parquet for Athena/Redshift) with crawlers/data catalog
• Schema evolution and data quality expertise (DynamicFrames, versioned schema pattern, nullable fields, validation alerts, dedupe, anomaly detection)
• Modernized legacy batch segmentation system into modular services (FastAPI + Glue/Athena + ClickHouse) using shadow-mode parity validation and incremental rollout
• SQL performance tuning with measured impact: reduced segmentation query runtime ~12s to sub-second and cut CPU ~40% via indexing and query rewrites
• Owned and resolved production data pipeline incident (CloudWatch/SNS detection, hotfix, backlog reprocessing, added drift-prevention controls)