QA Automation Engineer / SDET · AI agent tooling & evals · Málaga, Spain
I test software and build tooling for AI agents. For four years at Freepik (Magnific) I maintained the company's most active Cypress end-to-end suite and built the QA team's Claude Code setup (custom skills, safety hooks and a hierarchical memory system), which the whole team adopted.
Now I run a fully local coding agent (llama.cpp + opencode) and measure where local models can be trusted and where they make things up.
Projects
- llm-capability-evals: auto-graded evals for local LLMs (code against hidden tests, tool calls, agentic loops, honesty traps). No LLM-as-a-judge.
- local-agent-mcp-servers: web search and persistent memory MCP servers for local agents, in Python.
Tools I use every day
TypeScript · Python · Cypress · GitHub Actions · Docker · Model Context Protocol · llama.cpp · Claude Code
Interested in AI evaluation, QA for AI products and agent developer tooling. LinkedIn