About
Senior Backend Engineer with 20 years of experience building production systems. Currently focused on AI engineering — agents, RAG pipelines, evals, and LLM integration with legacy systems.
Based in Madrid. Working remotely.
What I build
I design and operate distributed systems that have LLMs as a core component: autonomous agents, RAG over internal data, eval loops, observability for AI in production, and integrations with heterogeneous legacy APIs.
My background is in event-driven architectures (Kafka, CQRS, Event Sourcing), high-throughput PHP/Swoole backends, and travel tech domain — a space where every provider has its own quirks, edge cases, and undocumented failure modes that no LLM will solve for you.
Stack
- AI / Agents: Anthropic SDK, Claude Code, tool use, RAG, pgvector, FastMCP, evals
- Backend: PHP 8.3, Python 3.12, Swoole / FrankenPHP, DDD, CQRS, Event Sourcing
- Data: PostgreSQL, ClickHouse, Kafka, Redis
- Infra: Docker, Kubernetes, OpenTelemetry, Sentry
Experience
Quadrant Travel Technology — Senior Backend Engineer & Tech Lead (2022 – present) Travel tech SaaS for agencies. Technical lead for the accommodations vertical. Integrations with Hotelbeds, Expedia, GoGlobal, Restel, W2M. Built the internal AI platform: Triage-Agent (Python, Anthropic SDK, production evals), RAG system over internal knowledge base (pgvector + sentence-transformers + MCP server).
Veloria Online Solutions — Co-founder & CTO (2014 – 2022) Built attribution and measurement software, REST/SOAP integrations with call centers, email marketing at scale. Full ops ownership.
Hispanetwork (2005 – 2014) Full stack development on high-traffic SEO portals. PHP, MySQL, JS.
Contact
- Email: javillav19@proton.me
- LinkedIn: linkedin.com/in/javiervv
- GitHub: github.com/jvillanueva-es