Antony G. - Al Testing and Prompt Engineer

Girona, Spain

Experience

Sep 2024 - Present

1 year 3 months

Girona, Spain

Developing Alethia AI: comprehensive platform for evaluating safety and alignment in LLMs using multi-judge evaluation strategies (LLM-as-judge methodology), enabling scalable and reliable AI safety assessments
Researching and establishing prompt engineering best practices, including systematic testing methodologies, optimization techniques, and reliability enhancement
Conduct comprehensive AI red teaming assessments across major LLM platforms (GPT-4, Claude, Gemini and DeepSeek), identifying safety vulnerabilities, bias patterns, and misalignment issues
Developed ModelArena: AI evaluation platform enabling comparative analysis of assistant responses with both automated and manual grading capabilities
Built QuickRecipeAI mobile application using GPT-4 Vision API, demonstrating practical implementation of multimodal AI systems
Created Q-Aware Labs as knowledge-sharing platform for AI quality assurance, prompt engineering, and safety research

May 2019 - Apr 2023

4 years

Barcelona, Spain

Performed detailed defect detection and manual testing across full software development lifecycle for medical software systems, ensuring compliance with quality standards
Reduced testing cycles by 20% through implementation of automated test pipelines using ALM-Jenkins integration and TestComplete, improving deployment velocity
Executed and maintained test automation scripts for SAP systems, covering critical business processes and reducing regression testing time
Configured integration between ALM Microfocus and TestComplete, creating seamless automated testing workflows
Developed customized testing solutions and reports for ALM Microfocus, enhancing test management capabilities
Collaborated with development teams to establish quality gates and testing protocols

Apr 2018 - May 2019

1 year 2 months

Girona, Spain

Provided comprehensive technical and functional support for point-of-sale (POS) software systems, maintaining 95%+ uptime
Managed and resolved technical incidents through systematic troubleshooting and problem-solving methodologies
Collaborated with end-users to identify, diagnose, and resolve software-related challenges, reducing ticket resolution time by 80%