QA Tech Lead with 13+ years in technology and emerging specialization in testing AI-powered systems: LLM chatbots, copilots, RAG architectures and Voice AI with ElevenLabs. I lead QA teams for enterprise clients across government, fintech, healthtech and e-commerce, building proprietary quality ecosystems from the ground up.
Creator of the QASL ecosystem: 12+ proprietary tools deployed in production and 30+ open-source repositories on GitHub. Direct leadership of teams up to 10 QA Engineers across simultaneous projects.
“I don’t just run tests — I build the systems others use to run them.”
Voice AI Stress Testing on enterprise conversational platform (client under NDA): Python wrapper + ElevenLabs SDK validating the voice→text→LLM→text→voice flow under load
Real-time QA Observability ecosystem with embedded AI — executive dashboard, installable mobile PWA, mobile-trigger execution and dynamic PDF report
1,300+ test cases generated with AI before development started (Shift-Left with static analysis)
614+ test cases executed on Multi-LLM platform · 92.9 ILT Score · 100% E2E Pass Rate · 0 critical vulnerabilities
Dual-VM QA Infrastructure with 29 tools approved by Management, Servers and SOC of the client
Currently building:voice-ai-testkit — open-source Python framework for stress testing and automated quality evaluation of Voice AI agents (ElevenLabs, OpenAI Realtime, Whisper). Upcoming release at github.com/E-Gregorio.
Published Methodology Framework
Proprietary operational framework · Published on GitHub under CC BY-NC-ND 4.0
Operational framework that unifies the practical application of public standards — ISTQB v4.0, ISO/IEC/IEEE 29119-3:2021, ISO/IEC/IEEE 29148:2018, OWASP LLM Top 10 2025 and NIST AI RMF 1.0 — into a cohesive, executable QA strategy. Includes operational templates, strategic guides, QA maturity model and the proprietary VCR framework (Value + Cost + Risk) for objective automation prioritization. End-to-end coverage: manual testing, automation, E2E, performance, security and AI/LLM testing. Repository: github.com/E-Gregorio/qa-shiftleft-methodology
EPIDATA Consulting • Government and enterprise sector • Buenos Aires • Hybrid
QA technical leadership across simultaneous enterprise projects. Design and implementation of the real-time QA Observability ecosystem with embedded AI, applied to the public sector.
Voice AI Stress Testing — enterprise project under NDA: Design and implementation of load testing on a conversational ElevenLabs endpoint validating the full voice→text→LLM→text→voice flow. Python wrapper + ElevenLabs SDK exposing an HTTPS endpoint, monitored with JMeter against the provider’s native dashboard. Saturation simulation (429 with fallback 500), latency validation, sustained-load behavior and conversational quality. Client and project under confidentiality.
Team leadership: Direct supervision of 10 QA Engineers distributed across 8+ simultaneous enterprise projects. OKRs/SMART definition, technical mentorship, deliverables review and quality control of the team’s work.
Real-time QA Observability ecosystem: Executive dashboard with embedded AI (Claude API), live environment monitor, installable iOS/Android mobile PWA, on-device execution trigger and dynamic executive PDF generator — all running in production with real data.
AI / LLM Testing: Complete framework with Garak (NVIDIA NeMo) for red-teaming, multi-model LLM-as-Judge (Claude/GPT-4/Gemini) with multi-dimensional evaluation, OWASP LLM Top 10 2025 coverage.
Static testing and Shift-Left: User Story analysis with AI (Claude API) for early gap detection and generation of 1,300+ test cases before development began.
Dual-VM QA Infrastructure: Design and deployment of a 29-tool architecture approved by Management, Servers and SOC of the client. Automation VM (Playwright, Newman, GitLab Runner) and performance/security VM (K6, JMeter, OWASP ZAP, Garak, MobSF) with Grafana/Prometheus/InfluxDB stack across 5 orchestrated Docker containers.
GitLab Runner Docker CI/CD pipeline: Stages INFO → E2E → API → NOTIFY → DEPLOY with automatic PASS/FAIL notifications and report publishing on GitLab Pages.
Advanced Performance Testing: Stress Testing (200–1,000+ VUs) and Soak Testing (1–4 hours) for race condition, memory leak and breakpoint detection under sustained load.
Mobile + Security Testing: Maestro for Android automation, MobSF for SAST of APKs/IPAs, OWASP ZAP for SQLi/XSS/CSRF on APIs.
QASL Ecosystem: Building of a suite with 12+ proprietary tools including NEXUS LLM (Multi-LLM platform with 12 microservices), QASL-TOMEX (multi-agent AI testing), INGRID v2.0 (LLM testing framework) and QASL-SENTINEL.
Automation engineer for multi-sector solutions focused on reliability, coverage and delivery velocity.
Design and implementation of Page Object Model frameworks in Playwright + TypeScript prioritizing maintainability at scale. E2E coverage across 4 simultaneous production products.
Load and stress testing with JMeter and K6 to validate performance and scalability across critical payment flows (PCI-DSS, ISO 8583) and authentication.
Test integration into GitHub Actions CI/CD pipelines, significantly reducing release feedback time.
Security testing with OWASP ZAP across regulated sectors (PCI-DSS for fintech, HIPAA for healthtech).
Mentorship and technical training for 15+ junior professionals in QA Automation and testing best practices.
Leadership of digital transformation projects in the healthcare sector: implementation of HIS (Hospital Information Systems) under HIPAA compliance across 5 medical centers, validation of sensitive clinical data migrations and coordination of multidisciplinary teams of up to 8 professionals.
QASL Ecosystem — Proprietary Tooling
Suite of 12+ tools deployed in production + 30+ open-source repositories. Full detail at e-gregorio.github.io/mi-portafolio
NEXUS LLM · Flagship · Multi-LLM Platform
Multi-LLM-driven QA platform. 12 microservices orchestrating Claude, GPT-4 and Gemini to automate the full testing cycle. 614+ test cases · 92.9 ILT Score · 100% E2E Pass Rate.
INGRID v2.0 · AI + Security Testing
Framework for LLM systems testing with LLM-as-Judge methodology, Red-Teaming via Garak (NVIDIA NeMo) and OWASP LLM Top 10 2025 coverage. Multi-dimensional evaluation: relevance, accuracy, coherence, completeness and hallucination score.
QASL-TOMEX · Multi-Agent AI Testing
Autonomous testing framework with 3 AI agents (Claude, GPT-4, Gemini) communicating via the QASL-CIPHER protocol for multi-perspective analysis and distributed decision-making.
QASL-MOBILE + Inspector · Mobile Testing Framework
Android mobile testing framework with an 8-level selector engine, Maestro for flow automation, MobSF for SAST and Claude Vision for autonomous UX/UI analysis.
QASL-SENTINEL-UNIFIED · Command Center
Command center with dual monitoring — Backend APIs (Prometheus) + Frontend DOM (Playwright + InfluxDB), 55 Grafana panels across 8 sections and a DEFCON classification system (DC-1 to DC-5).
QA Observability Ecosystem · Production · 2026
Executive dashboard (React 18 + Vite + AI) · Live environment monitor (proprietary Fullstack Scanner Dual Panel UI+API methodology) · Installable iOS/Android mobile PWA with on-device execution trigger · Dynamic executive PDF generator with real-time data.
Education
Claude Code in Action — Anthropic Academy (2025) · 87.5%
Playwright Advanced Testing — UPEX Academy (2023)
API Testing & Automation — UPEX Academy (2024)
ISTQB Foundation Level v4.0 — Planned 2026
Languages
Spanish — Native
English — Advanced written (technical reading, documentation, async communication). Intermediate spoken.