Experience

Software engineer with a track record in banking, insurance, retail and web product.

The full track record, in order. Where I led the technical decisions, what I built at each place, and how I have been putting AI agents into flows that were already in production.

  1. JIG
  2. Hiberus Digital
  3. Inditex
  4. Openbank, Santander Group
  5. VidaCaixa
  6. Foundation for Value Creation

Running right now

EQx - Elite Quality Index

July 2026 - Present

University of St. Gallen, Switzerland

Software engineer - EQx (Switzerland)

Foundation for Value Creation · EQx

Technical handover of the digital products of the Foundation for Value Creation: the public site of the Elite Quality Index and the private console behind it. The index measures across 151 countries whether elites create value or extract it, under the academic direction of the University of St. Gallen.

Home page of the Elite Quality Index, the index the foundation publishes
  • Handover from the previous developer with no interruption to the running product.
  • End-to-end ownership of the user experience across both products.
  • Direct contact with the index directors, designers and the people behind the rating model.
  • Product-level prioritisation: deciding what ships for each client milestone.
  • Remote work with an international team on an applied research project.
VidaCaixa

October 2025 - Present

Capgemini

Tech Lead - VidaCaixa

VidaCaixa · Capgemini

Frontend technical leadership on a strategic project in the insurance sector, defining architecture, development standards, best practices and the integration of generative AI into enterprise flows.

  • Frontend architecture definition and key technical decisions.
  • Leading the integration of generative AI across Caixa projects.
  • Agent orchestration, use-case assessment and the technical landing of AI flows.
  • Code reviews, quality standards and mentoring.
  • Coordination with backend, QA and business.
Openbank Santander Group

March 2025 - October 2025

Capgemini

Lead Software Engineer - Openbank

Openbank, Santander Group · Capgemini

Built the UI of a banking operations system for branches, focused on React, scalability, maintainability and hexagonal architecture patterns.

  • Building scalable banking interfaces.
  • Mentoring junior engineers on React and architecture.
  • Applying good practices on a large-scale financial project.
Minsait

June 2023 - March 2025

Inditex

Full Stack and Frontend Developer - Minsait (Indra)

Inditex · Minsait (Indra)

Development and maintenance of Inditex's store management terminal, deployed across thousands of points of sale worldwide.

  • Frontend with React and TypeScript; backend with Java and Spring Boot.
  • CI/CD with GitHub Actions, testing and quality improvements.
  • Scrum teams and deployments on cloud environments.
Hiberus

February 2022 - June 2023

React, Next.js, Node.js

Full Stack and Frontend React Developer - Hiberus Digital

Web applications in production, reusable components and projects such as the Hiberus corporate site and SivasDescalzo.

  • React, Next.js, Node.js, Jest and React Testing Library.
  • Components for the internal OnlyUI library.
  • Weather station management system as my final degree project.
JIG

September 2021 - December 2021

Wolfsburg mobility

Frontend Vue Developer (intern) - JIG

Frontend internship with Vue, CSS, HTML, Docker and Git, building a site for transport users and an admin panel.

  • Interface for bus line passengers.
  • Control panel for monitoring fleet components.
  • First professional contact with web product and real operations.

AI with a concrete job inside the product.

I work on agents, RAG, tools, MCP and LLMs from one simple idea: AI should solve concrete tasks inside the product and stay as maintainable as any other part of the system.

What interests me most is designing the flow: what context the model needs, which tools it can use, how responses are controlled and what experience the user actually gets.

VidaCaixa

generative capabilities integrated into an enterprise environment

Snowy

assistant and tooling over real weather data

principles

context, limits, traceability, cost and end experience

agent-runtime

01

Interface

input

02

Agent

reason

03

Tools

tools

04

Data

context

05

Control

guard

tool surface

RAG · MCP · AI SDK · evals · logs · fallback

lat

<1s

cost

cap

eval

on

How I approach it

Context before prompt

The value is in the model receiving the right information: your own data, permissions, retrieval, tools and the limits of the use case.

Agents with clear responsibility

I like designing flows where the agent knows what it can do, which tool to use and how to leave a trace you can debug.

Product, cost and operations

An AI feature also needs reasonable latency, fallback, observability, evaluations and a UX that doesn't force the user to understand the system.