I'm building an AI-assisted automation suite for Starscape 3.0 at Publicis Resources, using Playwright MCP, Playwright Agents, JavaScript, and GitHub Copilot Agent to expand web application automation and functional coverage.
Previously, I led QA strategy for AI-driven incident management at EY and designed end-to-end quality for a healthcare conversational voice AI agent at Deloitte. My work covered Smart Ticketing AI capabilities, ServiceNow ingestion, intent/NLU, STT/TTS, IVR integration, human-agent handoff, latency, hallucination, bias, jailbreak, and PII leakage testing.
At GlobalLogic, I architected QA frameworks for RAG chatbots with LangChain, FAISS, Hugging Face, LangSmith, Nemo Guardrails, and RAGAS. I also developed Python monitoring for inference latency, token consumption, CPU/memory utilization, and throughput, using evaluation metrics including MMLU, Perplexity, ARC, F1, and BLEU.
Across 14+ years in QA, I've delivered automation, performance, OTT, AdTech, healthcare, enterprise, telecom, and conversational AI quality programs. I've led teams and releases while building practical test strategies across web, API, mobile, TV, and AI platforms.
