Kaushik Sriram

About

Backend/data-platform engineer with Stripe and Salesforce experience focused on global payouts/treasury systems. Built an end-to-end payout settlement monitoring platform (FastAPI microservices, Kafka/Spark streaming, React dashboard, CloudWatch alerting) that cut settlement delays 25% and reconciliation time 30%, and productionized an ML anomaly detection service that reduced missed issues by 30%. Experienced modernizing monoliths into microservices with feature flags/canaries and close partnership with treasury/risk/CTO stakeholders.

Experience

Education

Key Strengths

• Owned end-to-end production feature across frontend, backend, data pipeline, and monitoring (global payout settlement monitoring)
• Built real-time streaming pipeline processing millions of settlement events/day (Kafka + Spark)
• Measurable ops impact: reduced payout settlement delays 25% and manual reconciliation time 30%
• Designed reliability mechanisms for event-driven systems (retries, exception handling, idempotency, out-of-order event handling)
• Production monitoring/alerting design with AWS CloudWatch and anomaly-based alerts; created runbooks for on-call
• Productionized ML anomaly detection service with evaluation (precision/recall) and guardrails; reduced missed issues by 30%
• Incremental legacy modernization with safe rollout strategies (feature flags, canary releases, parallel runs) and zero downtime/regressions
• Strong cross-functional partnership with treasury/risk/finance and CTO-level stakeholders; translated ambiguous goals into phased roadmap
• Effective embedded execution across multiple teams with minimal oversight; shipped in weeks via incremental delivery and async communication
• Used AI dev tools beyond autocomplete to accelerate refactors/debugging; reduced refactor time ~30%