Skip to content
πŸ‡¬πŸ‡§
Luciano Arguello

Luciano Arguello

AI-Driven Senior SDET Β· Playwright Β· LLM Evaluation Β· API & Backend Test Automation

CΓ³rdoba, Argentina πŸ‡¦πŸ‡· | Remote for US & Global Teams 🌎

Quality engineering, reimagined for the AI era.

I'm a Senior SDET, 15+ years building release pipelines that don't break under pressure β€” 1,000+ automated tests, 60% less manual work, 30% faster deployments at Upwork Inc., AppDirect, Inc., and Globant S.A. Today I focus on what comes next: LLM evaluation, agentic systems, and probabilistic validation for products where the right answer isn't always the same one twice.

Experience

  1. SDET β€” AI & Agentic Search

    May 2026 β€” Present Β· < 1 mo

    Upwork Inc. πŸ‡ΊπŸ‡Έ

    Defining quality strategy for AI-powered search and agentic systems: LLM evaluation, prompt regression, model evaluation, and layered deterministic + probabilistic validation for Upwork 2.0.

  2. Software Engineer in Test

    Oct 2025 β€” May 2026 Β· 7 mos

    Upwork Inc. πŸ‡ΊπŸ‡Έ

    Migrated legacy Ruby suites to modern Playwright e2e, designed scalable automation frameworks for backend + web, and drove shift-left practices across integration, API, and performance layers.

  3. QA Automation Engineer

    Apr 2018 β€” Oct 2025 Β· 7 yrs 6 mos

    Upwork Inc. πŸ‡ΊπŸ‡Έ

    Built the QA layer of the CI/CD pipeline (+30% deployment efficiency), validated 20 microservices over REST/GraphQL at 99.9% reliability, introduced Gatling load testing, and ran A/B tests on LLM-based search relevance that lifted match success by 15% for 500K+ users.

  4. QA Engineer

    Apr 2017 β€” Apr 2018 Β· 1 yr

    ThirdLove, Inc. πŸ‡ΊπŸ‡Έ

    Owned end-to-end test planning and reporting (–30% post-release defects), validated REST APIs with Postman, and automated UI flows via Ghost Inspector (–40% manual time, +25% coverage).

  5. QA Automation Engineer

    Sep 2015 β€” Apr 2017 Β· 1 yr 7 mos

    AppDirect, Inc. πŸ‡ΊπŸ‡Έ

    Automated the full legacy manual suite (–70% regression time), built 100+ UI & REST tests with TestNG + Selenium, managed Jenkins CI/CD (–50% deploy failures), and ran JMeter performance tuning (+20% perf).

  6. QA Analyst

    Mar 2014 β€” Sep 2015 Β· 1 yr 6 mos

    ARRIS Global Ltd. πŸ‡¬πŸ‡§

    Designed automation scripts for 50+ user stories (+40% process efficiency), validated REST APIs across 30+ stories (–25% integration issues), and shaped end-to-end testing strategy.

  7. QA Automation Engineer

    Aug 2011 β€” Mar 2014 Β· 2 yrs 7 mos

    Vates S.A. πŸ‡¦πŸ‡·

    Automated 80+ cases with JBehave + Selenium (–50% regression time) and executed 100+ manual tests (+30% defect detection).

  8. QA Test Analyst

    Nov 2009 β€” Jul 2011 Β· 1 yr 8 mos

    Globant S.A. πŸ‡¦πŸ‡·

    Led a 4-person QA team running daily pre-prod regressions; tracked 100+ bugs in Jira at a 90% pre-release resolution rate; coordinated daily with a US-based dev team.

Skills

  • LLM Evaluation
  • Agentic QA
  • AI-Driven Test Automation
  • Self-Healing Tests
  • Prompt & Model Regression Testing
  • n8n Workflow Automation
  • Playwright
  • Selenium
  • TypeScript
  • Python
  • Java
  • Cypress
  • REST API Testing
  • GraphQL Testing
  • Postman
  • Gatling
  • JMeter
  • k6
  • Jenkins
  • GitHub Actions
  • CI/CD Pipelines
  • AWS
  • Docker
  • Shift-Left Testing
  • Test Strategy & Planning
  • Microservices Testing
  • TestRail
  • Jira
  • Agile / Scrum

Certifications

  • Build AI Agents and Automate Workflows with n8n LinkedIn Learning
  • Software Testing Foundations: Keeping QA Skills Current in the Age of AI LinkedIn Learning
  • Software Testing Foundations: Integrating AI into the Quality Process LinkedIn Learning
  • AI for Software Testers: Apply AI Tools and Techniques to the Software Testing Life Cycle LinkedIn Learning

My Apps

Early Cat icon

Early Cat

Wake up Earlier. Win the day.

Coming soon
Early Cat screenshot

An iOS alarm app that gradually shifts your wake time earlier through science-backed circadian rhythm optimization. It forces you to scan a QR code placed in another room to dismiss the alarm β€” no snooze, no easy dismiss.

Get in Touch

Have a role in mind or want to talk shop about AI-driven QA? I'd love to hear from you.