

Results-driven Automation Test Lead and Quality Engineering professional with 13+ years of experience delivering large-scale digital transformation, banking, wealth management, fintech and energy programs across Australia. Proven expertise in leading automation transformation initiatives, building enterprise-grade automation frameworks, implementing shift-left quality practices and integrating automated testing into CI/CD pipelines. Strong hands-on experience with Playwright, Selenium, API testing, Tricentis Tosca, Azure DevOps and Agile delivery. Experienced in AI-assisted testing, LLM-driven test design, prompt engineering, AI-powered defect triage and intelligent test automation solutions. Recognized for driving quality engineering excellence, mentoring teams and reducing release risk while accelerating software delivery.
KEY ACHIEVEMENTS
RESPONSIBILITIES
Projects: NAB, EnergyAustralia
Project: Prosper
Prosper is America's first peer-to-peer lending marketplace, with more than 2 million members and over $6 billion in funded loans. It handles the servicing of the loan on behalf of the matched borrowers and investors.
Projects: MUFG Bank
MUFG is one of the world's leading financial groups. Headquartered in Tokyo. Confidential is a global network offering services including commercial banking, trust banking, securities, credit cards, consumer finance, asset management, and leasing.
· Designed and implemented E2E evaluation suites for LLM based AI agents using DeepEval, covering tool correctness, task completion, prompt alignment and step efficiency metrics.
· Built RAG pipeline evaluation covering hallucination detection(Faithfulness metric), retrieval ranking quality(Contextual precision) and retrieval coverage(Contextual recall) using GPT-4o as judge LLM.
· Evaluated multi turn chatbot conversations using ConversationalTestCase with KnowledgeRetention, TurnRelevancy and ConversationalCompleteness metrics to catch memory failures and role breaks across turns.
· Implemented safety evaluation suite covering Bias, Toxicity and PII leakage on a live support agent using DeepEval's synthesizer to auto-generate test inputs directly from policy documents.
· Instrumented LangChain agents with DeepEval tracing(@observe, CallbackHandler, update_current_trace) to capture tool calls, LLM spans and retrieval context for metric evaluation without modifying agent logic.
· Authored custom evaluation criteria using GEval and conversationalGEval to assess domain-specific quality dimensions beyond pre-built metrics.
Australian Citizen