Rhanny Belle Urbis
AI Automation Engineer | Backend Systems & AI Agents
I build efficient systems.
I build backend services, AI workflows, and automations that help teams reduce manual work, run reliably, and handle business data securely.
Technical skills
Technical Skills
I use these tools to build, test, secure, deploy, and maintain backend systems, AI workflows, and business automations.
Backend Systems
PythonJavaScriptTypeScriptFastAPIPydanticREST APIsSwagger
AI & Workflow Automation
LangChainLangGraphLlamaIndexRAGLLM API integrationn8nMakePower AutomateOpenWorkflow
Databases & Integrations
PostgreSQLMySQLSQLiteSupabasepgvectorMilvusRedisMicrosoft Graph APIGoogle Workspace
Security, Testing & Reliability
End-to-end testingAPI testingQA automationDASTWeb application testingMonitoringFailure handling
Cloud & Deployment
DockerAWSAzureGoogle CloudRenderVercel
Frontend & App Interfaces
ReactNext.jsSvelteKitHTML5CSS3Tailwind CSSStreamlit
CRM & Business Platforms
HubSpotWelcomeHomeLineLeaderZendeskFreshserviceRingCentral
Selected work
AI/Agentic AI projects
Here are a few apps and systems I’ve built around AI interaction, information retrieval, and workflow automation.
TripBites
A travel assistant built with Next.js and LangGraph. It uses a multi-step workflow to gather trip information and return a useful answer in one place.
TypeScriptNext.jsLangGraphReAct
View project →VoiceBuddy
A document-to-audio app that summarizes uploaded documents and turns the summary—or the full document—into narration using OpenAI TTS.
PythonStreamlitOpenAI TTS
View project →CarInsure Bot
A question-and-answer app for car-insurance documents. It retrieves relevant policy information before answering questions in plain language.
PythonStreamlitLlamaIndex
View project →Writing
Engineering articles
Selected articles focused on retrieval design, backend reliability, practical architecture, and workflow automation.
Embeddings vs. Re-Ranking: How to Use Each
When embeddings are enough, when re-ranking is justified, and how to combine both in production retrieval pipelines.
RAGRetrievalEmbeddingsRe-ranking
Read on Medium →How to Build a Fast & Reliable API for an ETL/Report Generation Agent
A practical pattern for keeping FastAPI thin while moving heavy ETL and report generation work off the request path.
FastAPIETLAgentsArchitecture
Read on Medium →Building Production-Ready Apps Without Over-Engineering
How to keep systems dependable and maintainable without adding unnecessary complexity too early.
ArchitectureBackendProduction
Read on Medium →Why Most Systems Don’t Have Real Error Handling
A production-minded look at failure handling, stability, and what real operational resilience requires.
ReliabilityError HandlingBackend
Read on Medium →