Description

Ingeniero de software senior (17 años) especializado en llevar sistemas de IA a producción de verdad, no a demos. Diseño y opero aplicaciones LLM y sus pipelines de datos en Python/FastAPI sobre AWS/GCP: RAG, orquestación de agentes con MCP, streaming y analítica de llamadas, con observabilidad completa (coste, latencia, trazabilidad), Docker/Kubernetes y CI/CD.

Ahora mismo tengo 11 agentes LLM en producción. Algunos resultados: coste de voz-LLM a 0,04 $/min enrutando a modelos open-source, −35% de latencia y −30% de cómputo en flujos críticos, y ROAS de Meta Ads del 4% al 9% con un agente MCP.

Me gusta mancharme las manos con el código. Trabajo AI-first y spec-driven (TDD, contract testing, una decisión documentada por cada cambio) y cuido que el software sea testeable, escalable y barato de operar. Fuerte en APIs: contratos versionados, OpenAPI, webhooks, microservicios. Perfil full-stack y de data engineering (Python, FastAPI, TypeScript, Next.js, ETL/ELT, Spark, Flink, BigQuery) sobre cloud (AWS, GCP).

Busco un rol senior como individual contributor, por contrato o largo plazo, 100% remoto (cualquier huso) u on-site en Madrid.

Versión aún más corta por si el campo es breve:

Ingeniero senior (17 años) que lleva IA a producción de verdad: aplicaciones LLM y pipelines de datos en Python/FastAPI sobre AWS/GCP (RAG, agentes con MCP, streaming) con observabilidad, Docker/Kubernetes y CI/CD. 11 agentes LLM en producción; voz-LLM a 0,04 $/min, −35% latencia, −30% cómputo, ROAS 4%→9%. Hands-on, AI-first y spec-driven (TDD, contract testing), fuerte en APIs y microservicios. Busco rol senior IC, contrato o largo plazo, remoto u on-site en Madrid.

¿Te sirve, o la quieres con un tope concreto de caracteres?

Industry field of expertise

Languages

Spanish
Native or bilingual
English
Native or bilingual
Portuguese
Fluent

Workplace preferences

Can work on-site

Madrid (up to 50km)

Independent (B2B via own company)
Senior AI / Backend Engineer
January 2023 - Today (3 years and 6 months)
— Build and operate Python / FastAPI microservices and JS / TS clients on versioned API contracts; LLM agents and data pipelines in production (OpenAI
• · Anthropic
◦ · MCP
◦ · AWS / GCP).
— Built pilis.app: production LLM agent API (OpenAPI/Swagger) plus a data & analytics layer — post-call analysis pipelines, model management and internal/external dashboards (incl. Meta Ads) — consumed by a Next.js web and a React Native app on the same contracts; CI-gated TDD on every critical flow.
— Technical leadership: migrations, safe refactors under coverage, observability, and ramp-up of junior teams.
LangGraph FastAPI MCP
Migro
Founder & Full-Stack Engineer
January 2022 - January 2025 (3 years)
Legal-tech platform for immigration processes in Spain
• · Remote
◦ · owned product & architecture end-to-end
— Designed and ran 11 LLM agents in production (MCP + LangChain): legal research, case management, paid acquisition, voice analysis, compliance, document processing and governance.
— Built a voice-LLM call-center (ElevenLabs → Telnyx) writing structured CRM records; cut cost to $0.04 / min by routing to open-source LLMs.
— Meta Ads MCP agent operating paid acquisition: ROAS 4% → 9%, CPL −24%; clean API layering delivering −35% latency and −30% compute spend.
MCP Desarrollo Back-End FastAPI Docker
Uber
Expansion & Driver-Acquisition Manager
January 2015 - January 2018 (3 years)
Lima, Peru
(Peru, Colombia, Ecuador)
— Owned driver-acquisition strategy, onboarding flows and data-led growth for multiple city launches; cross-functional work across ops, marketing and expansion.
— First hands-on with data engineering here — Spark and Flink for ETL/ELT and BigQuery pipelines on acquisition data — which sparked the pivot toward building data systems.
— Negotiated a lead-acquisition sharing agreement with a top advertising agency (2016) — ~$2M saved in acquisition spend.