Florian Krueger

Delivered a data-driven alerting web app (React/Next.js + Express / Python), secured with Amazon Cognito and API Gateway; containerized services on EC2 with real-time notifications via Socket.IO for clinician workflows. Designed Python task schedulers and async pipelines to process high-volume Meltwater social streams, maintaining stable ingest latency under load. Built a real-time analytics dashboard (FastAPI + React/Next.js) with interactive drill-downs for rapid triage during spikes. Implemented AI-driven sentiment and intent analysis with AWS Bedrock and GPT-4 (guardrails; batch/stream modes) to improve signal quality and reduce alert noise. Modeled data across PostgreSQL (RDS) and DynamoDB, and deployed analytics microservices on GCP for batch/stream workloads. Established CI/CD with Jenkins and GitHub Actions; added observability and SLOs; collaborated with Data Science on feature stores and evaluation harnesses.

Led frontend architecture for real-time analytics and data visualization platforms used by internal product teams and enterprise customers; built TS + React/Next.js UIs with reusable component libraries and D3/Recharts, improving page load (p75 TTI) and increasing weekly active users. Implemented robust state management with Redux Toolkit/RTK Query and Apollo Client (GraphQL); introduced code-splitting, lazy loading, and HTTP caching to keep Core Web Vitals in the good range under peak traffic. Designed and delivered scalable backends in Python (FastAPI) and Node.js, containerized with Docker and deployed on AKS; added async workers (Celery/Redis) and streaming endpoints (Server-Sent Events/SignalR) for live dashboards with sub-second updates. Drove enterprise digital-asset accounting pilot integration (Azure -> QuickBooks Online) to improve reconciliation cycle time and scale transactions. Modeled data (PostgreSQL/Cosmos DB) with targeted indexes and read/write segregation; implemente

Developed a high-performance marketing/commerce web platform; built a modular design system and reusable UI components (React JS) with Storybook, reducing feature lead time and improving maintainability. Architected REST and asynchronous APIs using Django and FastAPI; introduced background jobs and pagination/streaming to cut P95 latency. Integrated React frontend with Django/FastAPI backends via optimized endpoints, JWT auth, and SSE/WebSocket for real-time updates. Modeled relational and document data (PostgreSQL, MongoDB) with proper indexing and read replicas to reduce latency. Led performance/load testing (Locust/pytest) and set SLOs/dashboards; automated inventory + order-tracking workflows; deployed via CI/CD to GCP (GKE/Cloud Run) with canary releases and feature flags. Collaborated with cross-functional teams to mentor on API contracts, React performance, and DB optimization.