Piotr Nowak

Architected an event-driven processing platform handling 2B+ events per day across three AWS regions, improving throughput and resilience through partitioning, backpressure, idempotent consumers, and automated failover. Led decomposition of a monolith into 30+ independently deployable services, reducing deployment lead time from 5 days to under 1 day and enabling independent team releases. Redesigned API and data-access architecture to reduce p99 latency from 480ms to 95ms using query optimization, Redis caching, connection-pool tuning, and targeted service decomposition. Designed Kafka-based workflows with retry, deduplication, dead-letter, and replay mechanisms, reducing failed-message recovery time by 80%. Migrated production workloads to Kubernetes on AWS with Terraform-managed infrastructure, progressive rollouts, and standardized rollback procedures. Established organization-wide observability standards using OpenTelemetry, Prometheus, and Grafana, reducing mean time to detection

Designed and implemented Python and TypeScript microservices for customer-facing workflows handling 50M+ requests per month. Migrated a legacy synchronous workflow to an asynchronous event-driven architecture using Kafka and Redis, increasing throughput by 4× and reducing timeout-related failures by 62%. Built versioned REST and GraphQL APIs for 20+ internal/external consumers, establishing reusable patterns for authentication, validation, error handling, and observability. Optimized PostgreSQL schemas and queries, reducing a critical reporting workflow from 14 minutes to under 2 minutes. Introduced automated integration and API testing, increasing critical workflow coverage from 35% to 82% and reducing production regressions. Automated AWS infrastructure provisioning with Terraform and standardized CI/CD using GitHub Actions, reducing new environment setup from 2 days to less than 4 hours. Participated in on-call rotations and led root-cause investigations for customer-facing product

Built backend services and customer-facing applications using Python, Node.js, React, PostgreSQL, and TypeScript. Designed REST APIs and asynchronous processing workflows supporting third-party integrations and internal business systems. Optimized PostgreSQL queries and database access patterns, reducing report-generation time by 45%. Built reusable React/TypeScript UI components adopted across multiple applications. Introduced automated unit and integration testing for critical backend workflows to improve regression detection before releases. Participated in architecture discussions, code reviews, and production troubleshooting across the full software lifecycle using Agile delivery.