You like adventures, talking around a fireplace while roasting marshmallows.
Without any doubt, you are a qualitative-focused person, choosing quality over quantity. The world needs more beings like you. Great choice!
You want situations, tasks, actions and results: numbers and ecstasy.
You don’t like to roast marshmallows, you just want to eat the marshmallow.
Only got 20 seconds? This mode’s for you. Save the deep, life-changing story for when you’ve got time to dive in.
The marshmallow you just named is that dot down there. We will follow Dotty throughout this page, as it stands out from the crowd.
At each step of my life, as a professional and as a human, I met someone like Dotty: somebody at a moment of choice, who needed something to become their best version.
Economics, statistics, sociology… lots of disciplines make Dotty fit in a model: another dot within the Gaussian bell. Dotty is standardly unique, and will act depending on the context, information, and surrounding objects. I was fascinated by how the sum of individual people like Dotty creates the global society.
I research, study, and learn about Dotty: what does Dotty like? Why? What are the major inconveniences Dotty faces? How can they be fixed? Is the resulting design useful for Dotty? Will it be useful for other people?
Dotty is surrounded by a few favorite daily objects. These objects talk about Dotty and help me understand Dotty’s interests.
These objects are usually given by context and society, considering Dotty’s economy, situation and relationships.
Originally, I designed interfaces, wireframes, and devices. Always thinking of Dotty and the context around Dotty.
Globulus (2018)
But somehow, a person like Dotty was always behind each interaction, wireframe or prototype.
I designed the user experience (UX) and user interface (UI) for open.unicef.org with Dotty’s needs in mind, focusing on making detailed information about $3.02 billion in funding easy to find and understand.
Read the caseDotty needed information tailored to different circumstances, so I designed a customizable dashboard where Dotty could pin, visualize, compare, and export the relevant data.
Read the caseWith responsive charts and visualizations, Dotty could explore the data and uncover insights within a customizable dashboard.
Read the case
Just like the 20,000+ users of the Patient Portal at Sant Joan de Déu Children’s Hospital in Barcelona, the main center for rare pediatric diseases in Europe and a worldwide reference.
Read the caseI led the design process so parents like Dotty could use the Portal and communicate better with the hospital.
Read the case
The redesign improved the NPS by 5 points (from 76 to 81), meaning that the average user now proactively promotes the hospital portal.
Read the caseA crucial parameter for the perception of quality during a doctor’s visit is the same one as love: the amount of time spent looking into somebody’s eyes.
To increase this number, we designed and tested the use of this technology during medical visits.
Read the caseLaia (Lilly, in Catalan) is an AI scribe that was designed and tested with users at every stage of the project. The experience during the visit is seamless, and families can open up more to their clinicians.
Read the case
The doctor listens during almost the whole visit (95.7%, precisely) and makes direct visual contact, while reducing the amount of work and documentation. Beyond a wide implementation, the output has also been an academic paper, so other hospitals know the advantages of this kind of technology.
Read the case
Due to over-information of data and the overwhelm that big numbers represent, audiences can lose the link between data and reality.
Read the caseThis project began as an XR exhibition in Barcelona, featuring Tinder conversations between people affected by war. It evolved into an AI agent that portrays a person living in a conflict zone, allowing anyone to ask questions about what life is like amid violent conflict.
Read the case
The AI conversation was designed with current events in mind. Originally released in 2020 and awarded a Laus in 2021, the project has since been reinvented and is now accessible everywhere.
Read the caseSince 2019 I’ve taught design, UX and AI at different Barcelona universities, at the meeting point of these three focuses that, in the end, make me work for Dotty.
You followed Dotty from the crowd all the way to the fire. Roasted marshmallows taste better, right?
I’m glad you finished the story mode and read about the things I have been doing.
However, the most exciting ones are the new ones we can do together.
Are you ready to work?
This is a postcard souvenir from the story about Dotty: proof that you followed the dot from the crowd to the roasted marshmallow. Turn it over, keep it, or send it back to me.
Direct mode · 20 seconds
Senior UX, product and interaction designer in Barcelona. 8+ years across clinical AI, conversational products, transactional flows and complex data. Economist by training. Open to Europe and US relocation.
Before design
Economics, statistics and how individual decisions add up to a society.
Design
Interfaces, wireframes and devices, from Globulus (2018) to work for Zara, Iberia, Santander and more.
Teaching
Design, UX and AI at Barcelona universities since 2019.
Global-to-detail Personalized Data Visualizations
$3.02B: of humanitarian aid funds that can be audited
Situation and Tasks
The UNICEF Open Data platform holds a massive amount of information. Users may find it difficult to navigate through the different multi-level views, from global to detailed, and therefore understand the real impact of programs and donations.
UNICEF commissioned Domestic Data Streamers, which brought in Galactic Creative Studio (where I was Creative Director and lead UX/UI designer) to design the site’s UX and UI.
The main issue was the repetitive flow users had to follow to get multiple data sets. Users had to navigate multi-level views from global to detail, which made the real impact of programs and donations hard to see and compare.
With all the project stakeholders, we defined the project’s main purposes. We co-created a new flow that lets the user cross-reference the data sets and visualizations needed. Now, the user can create their own dashboard with the information they need, download all the data sets, and share them with others without having to log in. Our goal was to translate this flow into pixels.
Actions
UNICEF had already identified its users and their main goals. Those users were mainly donors, country stakeholders, and researchers. Each had a different data goal for the platform. The challenge was to identify and deliver the data that each one wanted, without adding visual noise. With simple dropdown lists, the users could select the data they wanted to cross and build their own board.
Actions
The main focus for this platform was user personalization and the global-to-detail navigation structure. We needed to create an easy-to-use tool for navigating data, but also one that allowed on-screen data visualizations.
We benchmarked board-based products, including social platforms, and chose a visualization tool with a comprehensive navigation architecture that allowed us to pin and save the most relevant visualizations.
Design Challenge
The challenge was to build multiple data visualizations that could adapt to different window sizes and responsive layouts, while remaining accessible and relevant to each user.
Color palettes and data visualizations are chosen based on UNICEF brand guidelines, the number of variables, and the purpose of the visualization.
Users can create their own data combination and export it to CSV or XLS files. They can also download their visualizations combined as a PDF.
The data visualization adapts to different devices and window sizes, ensuring everyone can access the data, even on obsolete devices.
The colors in the original UNICEF palette are too bright and lack WCAG AA contrast ratios. Therefore, an accessibility mode was designed to modify those colors and optimize legibility.
Results and Impact
The new visuals were well received, allowing auditing of $3.02 billion in humanitarian aid funds, detailing the spending on 448 emergencies in 104 countries.
Although there is no specific impact data for the published update, its visual accessibility and the navigation to obtain personalized data helped to attract a community of more than 300 collaborators on GitHub, who interact with the open data platform.
Reducing the gap between user needs and digital healthcare communication
76→81: Patient Portal NPS after the redesign
Situation and Tasks
The Patient Portal is the primary communication support channel of Sant Joan de Déu Children’s Hospital for 20,000+ patients and families, handling 1:1 clinical messaging, self-service information, and privacy/GDPR flows, and gradually replacing SMS and the call center. I owned its interaction design and the Voice-of-Patient feedback loop.
Since its pre-release, I have been working on continuous validation, testing, and iterative improvements to enhance core functionalities, including messaging the clinical team, privacy and GDPR applications, and clinical information sharing, among others.
Actions and Methodology
To understand Portal usage at a user experience level, three main techniques are used.
Focus groups helped families share their thoughts and areas of improvement. These discussions help prioritize needs and understand which functionality improvements are more urgent.
Through user flows, wireframes, and high-fidelity prototypes, I validated how they use the Portal. The goal was to test its functionalities and design updates before developing them. Tests were conducted across waiting areas, reaching families without requiring formal recruitment and enabling agile iteration and optimization.
Qualtrics surveys about overall and specific functionality satisfaction are designed in collaboration with the Voice of Patient area. These surveys help to evaluate overall functionality and identify the next strategic steps.
Results and Impact
The Patient Portal is a crucial part of the hospital experience. More than 20,000 users have registered, and 5,000 are heavy users, most of whom have complex conditions that require frequent hospital visits.
Considering the 2024 launch, the impact on design and information architecture was positive, increasing the Patient Portal NPS from 76 to 81. The Portal has reached 60% adoption, with 90% projected by 2027.
Future
The expectations of the Patient Portal are solid. Considering the cost-optimization factor, its use as a communication channel will gain ground over other channels, such as SMS or the telephone contact center. Since more management options can be added to the Portal, users can easily access information on how to prepare for the visit. This implies making the Portal available on wearable devices to create continuity between digital channels, space, and the healthcare model experience.
However, there are barriers to overcome. For example, developing strategies to address digital poverty, by providing alternative access methods or digital literacy support. Another blocker is professional resistance. It is necessary to promote Portal engagement to healthcare teams as the exclusive communication tool with patients and families.
Measuring the human impact of AI during medical visits
95.7%: of visit time with direct eye contact, up from 78.6%
Situation and Tasks
Sant Joan de Déu Children’s Hospital piloted Laia, an AI scribe that transcribes the consultation and drafts the clinical note, in pediatric outpatient clinics.
I led research, interaction design, and implementation management for Laia, assessing whether AI improved patient-family interactions and overall visit quality, considering the interaction between clinician and family, and the quality of what ended up written down.
Six months, 40 clinicians across hematology, rheumatology and general pediatrics, 197 documented visits. The results were published in Medicina Clínica, where I figured as the second author.
Actions
The design process starts with collaboration among the team of professionals who will use Laia and the engineers who will prepare the pilot test. During a co-creative session, we determine the user flow and the interactions to make its use non-invasive during the whole visit; the technical constraints, and how the data needs to be managed to protect patient’s privacy.
Laia’s user flow is adapted to focus on clinicians’ work and documentation tasks. The interactions are positioned to cause the least friction between users and professionals.
When a new recording starts, a simple screen with information related to how to use it, and how to tell patients that the conversation is being recorded, and it is their right to stop the recording at any moment.
Laia uses a simple animation with the same color palette as the hospital branding, indicating slow movement while recordings are being made.
A simple layout with the main stages of the visit, with pre-filled text fields with the information Laia understood, organizes the clinical information. On this screen, the professional can correct any errors and validate the information. Once finished, the whole text is copied and ready to paste into the Integrated Clinical History Software.
Actions
To measure Laia’s impact on visit quality, the research combined three strands.
The interviews explored how families felt about AI being present during the visit, what they wanted to be told beforehand, and their privacy concerns. AI-generated storyboards were used to make the scenario concrete.
Each segment of the consultation was defined and timed: case history, examination, results, next steps, family comments, direct eye contact, and time spent on the computer. These definitions made the comparison possible.
Thirty timed observations, fifteen with Laia and fifteen without. Separately, three pediatricians blind to the condition scored thirty clinical notes against a rubric based on the validated mPDQI-9.
Results and Impact
Direct eye contact with the child and family rose from a median 78.6% of visit time to 95.7% (p<0.001). Families stated the clinician was more present, and clinicians described being freed from typing while trying to listen. Visits ran 25.7 minutes against 21.9, a difference that did not reach significance, so Laia did not measurably lengthen the consultation.
The notes changed too. Overall documentation quality rose from 17 to 23 out of 25 (p=0.008). The clearest single finding: notes written with Laia recorded what had actually been explained to the family in 80% of cases, against 20% without it (p=0.003). Clinicians tend to omit what feels obvious in the room. That omission is exactly what matters months later.
Concerns and future
The consultation improved. Adoption did not follow. Only 12 of the 40 trained clinicians kept using Laia, and six never used it at all. Two-thirds hit technical problems, most often information landing in the wrong field and transcription errors on drug names and specialist vocabulary. Reviewing and correcting each note took 33% to 38% of the visit time.
While patients and families welcomed Laia, the key to widespread adoption lies with professionals. Strengthening Laia’s vocabulary and comprehension will reduce professionals’ reluctance and will encourage its use during medical visits.
Humanizing war data through an XR and an LLM agent experience
3,000: people in conflict areas willing to share their story
Situation and Tasks
One death is a tragedy. A million deaths is a statistic. This concept, also known as Compassion Fade, describes how human empathy diminishes as the scale of suffering grows. This project aims to reduce Compassion Fade related to the large, abstract, and distant numbers of war-caused casualties.
Its execution lasted 6 months: from February 2020 to July 2020. During that period of time, most of the news showed COVID-19 deaths in thousands. Since the situation directly appealed to individuals, the Compassion Fade perception was absent, as there was strong alignment with the closest reality. However, would the Compassion Fade come back again? Can we make large numbers approachable? War Owned approaches this question by experience rather than argument.
In 2026, wars are still happening. The project again revisits the topic of communicating war data by building an agent on the Claude API, with prompt and dialogue design, an interaction model, visual language, and front-end implementation, all owned end-to-end.
Actions
The project draws on the Identifiable Victim Effect. This psychological principle arises from the fact that, although humans have two coexisting ways of thinking (rational thinking, grounded in data and facts, and experiential thinking, grounded in emotions), numerous researchers agree that experiential thinking has more impact on behavior, which leads to the Singularity Effect: people identify with and emotionally relate to a single individual rather than a group or large number of people.
The Singularity Effect, combined with the Vividness Factor (which suggests that individuals respond better to specific interactions and graphic information), served as the main theoretical framework for this project. Combined with the main conflict data from ongoing wars in Syria, Yemen, Afghanistan, or the Cartel wars of Mexico, this framework pointed toward a design to diminish Compassion Fade: an experience to make individual stories visible within statistics.
Actions
Applying the Identifiable Victim Effect was also required. Therefore, it was necessary to connect with people living in a conflict area who were willing to share their stories to conduct qualitative interviews. Due to the emotionally relatable individuality of the singular effect, the chosen platform was Tinder, especially because it offered the “passport function”.
Given that most of the conflict areas were in countries with legislation against same-sex relationships, and there is evidence of higher male engagement on the platform, a female research profile was created, detailing the purpose of the study and actively asking for consent before proceeding to interview. All the collected data would be anonymized.
The response was extraordinary: more than 3,000 users expressed their willingness to participate. With these users, 30 chat-like conversations were conducted. The main insight was that each had similar affections, but each had its own narrative that created a genuine connection. This powerful insight served as the basis for developing the project’s design solution.
Results and Impact: XR
The design solution required a specific setting that would help the audience connect instantly with the narratives collected.
In Barcelona’s Gothic district, there is a small and quiet square where there are still shrapnel marks from a bomb that exploded during the civil war, almost 90 years ago: “Plaça Felip Neri.” Today, people gather here to have intimate conversations, without considering that this square was a conflict zone. This historical place was the perfect location to host the exhibition of war human narratives.
The format adds a living layer on the spot, giving evidence that, although visitors are not physically in a war zone, others are currently living in it. Extended Reality was the most appropriate format to express it for three different reasons: it erases the distance between visitors and active conflicts directly in a place where reality shows the marks of war; it addresses the Vividness Factor through live conversations with people who are living in a conflict area; and finally, it combines the experiential layer with general statistics of the conflict, engaging both rational and experiential thinking at the same time.
Results and Impact: AI
The exhibition ran for three weeks at Plaça Felip Neri, welcoming approximately 530 visitors. Furthermore, the project was awarded the Laus Award for Digital Design by the Spanish Association of Graphic Design in 2021.
As global conflict has intensified, War Owned has evolved into a new experience that applies the same design principles: address the Vividness Factor, explore general statistics about conflict, and diminish the distance individuals feel from it through AI.
The Ukraine conflict has been going on for more than 1,500 days. However, since people are used to it, they may feel desensitized when hearing news about this war. To increase accountability for this situation, the project offers a custom AI agent via the Claude API and a persona profile derived from 10 qualitative interviews with Ukrainian families. You can check the results on this webpage: war-owned.com.
One death is a tragedy. A million deaths is a statistic. But somehow, LLM statistics can be a great tool for transforming a large number of people back into a single person, entering a new paradigm where it becomes possible to humanize large statistics and, therefore, dissolve the Compassion Fade.
I’m a Senior UX, Product and Interaction Designer with more than 8 years of experience. I studied economics, interaction design, and data visualization.
My main interest is understanding human behavior, as specific knowledge delivered at the right time can modify actions or purposes.
I believe everything can be related to design, from specific daily interactions to global data.
Work
Research
Studies
Teaching
Speaking
Recognition
