Luca Craciun

I’m an AI Automation Engineer specializing in reliable multi-agent LLM systems, AI workflow automation, and backend infrastructure for production use. I focus on building quality-control frameworks, RAG pipelines, evaluation workflows, and automation-ready APIs.

At Genpact, I delivered measurable operational impact by automating complex reporting processes, reducing monthly manual effort by ~67 hours. I designed and implemented a reporting automation using Google Apps Script and SQL, cutting a 100-agent monthly cycle from 20–40 minutes per report to under 5 seconds.

I also strengthened operational reliability through systematic Root Cause Analysis (RCA), eliminating recurring failures and improving process reliability. Alongside this, I developed KPI dashboards and real-time reporting pipelines used by operational leadership for performance tracking and decision-making.

I enjoy turning repetitive workflows into structured, agentic processes—supported by strong engineering practices like structured logging, analytics dashboards, and CI/CD. My projects reflect a focus on prompt engineering, hallucination reduction, safety-first triage, and end-to-end retrieval with citation-friendly outputs.