Cloud & AI/ML engineer building agentic workflows, RAG systems, and MCP tooling on AWS.
- Portfolio site: oscar-valles.com
- Blog: Medium @ovalles6845
- M.S. Computer Engineering (cloud & AI focus), @ University of Texas at Dallas
| Project | What it demonstrates |
|---|---|
| raditriage · live demo | Agentic veterinary radiology workflow — classify→RAG→draft→route pipeline, RBAC, SLA dashboard, key-safe Claude proxy |
| cloud-sre-agent · live demo | Bedrock SRE/FinOps agent — CloudWatch, Cost Explorer, Trusted Advisor, HITL remediation |
| agentic-governance-harness · live demo | Agentic AI governance benchmark — guardrail trip rate, policy latency, cost per resolved task |
| agentic-ai-workflows | Single & multi-agent patterns with LangChain, LangGraph, CrewAI, AutoGen |
| pgvector-rag-demo | RAG with PostgreSQL + pgvector, local embeddings, Streamlit UI |
| aws-cloudops-mcp | AWS CloudOps MCP server (EC2, CloudWatch, S3, Lambda, …) |
| small-biz-operations-mcp | MCP agent for small-business ops + dashboard |
| ml-experiment-tracker-mcp | ML experiment tracking MCP server (SQLite) |
| budget-insights-platform | Cloud-native finance platform (K8s, AWS microservices) |
| fraud-detection-system | Flask API + React dashboard for transport fraud analytics |
| ml-cloud-deployment | Spam detection ML → AWS Lambda + API Gateway |

