Hospital Sant Joan de Déu is one of Barcelona’s leading hospitals. Its Learning by Doing program aims to build AI awareness by enabling participants to developreal solutions for real needs while learning how to use AI along the way.
We were one of the consulting firms selected to participate in this program, and we partnered with DRIA (Direcció de Recerca, Investigació i Aprenentatge), the unit responsible for research and professional training, to help them build a solution to improve the analysis of training needs.
The challenge
The process for detecting learning needs among professionals in the hospital was extremely manual. They collected answers through a survey in which specific training actions were requested.
Cleaning, review, classification and grouping required an intensive human intervention. The resulting Excel spreadsheet required a lengthy and complex manual process that could require months.
The goal was to transform dispersed and varied answers into structured information that could be compared to support better training planning, without eliminating the expert unit’s review and analysis.
What we did
Working with the DRIA expert team, we designed and validated an MVP to automate the process, conceived as the first step in a larger architecture that could later be developed:
- Capturing the information in a structured flow
- Processing the answers with AI
- Identifying real training needs present in the answers
- Normalizing, classifying and grouping those needs
- Applying scoring and prioritization criteria
- Generating provisional results for human review
- Incorporating a human validation step before consolidating results
- Producing prioritized lists, dashboards and reports that are useful for the DRIA team
We structured this solution in five layers:
- Data collection
- Processing with AI
- Storage and traceability
- Human review
- Output generation
And, for the processing stage, we designed an AI pipeline, orchestrated with Power Automate, with three steps:
1. Normalization and classification
Initial outputs become structured information. Our system identifies the need, generates and classifies a normalized output, and determines the service and field involved.
2. Semantic grouping
Similar training needs could be expressed differently in different answers. This step of our pipeline groups them when they express the same underlying learning need, reducing duplications.
3. Scoring and prioritization
The needs that are already grouped in the previous step are evaluated using defined criteria. The outcome is a prioritized list for deciding which needs should be considered in the next training planification for professionals.
The flow works as follows:
- A file is uploaded
- A first step of the AI pipeline is executed
- An intermediate result is generated
- Human validation is requested
- If approved, the system moves forward with semantic grouping and prioritization
- The final result is stored in SharePoint, and the team is notified of the output through Teams
What we achieved
The MVP provides a robust foundation for a more efficient and traceable process, reducing manual work and applying consistent criteria to the evaluation.
Our project received the “Flywheel” award in the Learning by Doing presentation, which recognizes its ability to establish a continuous improvement cycle.
As for next steps, our plan included two further developments:
2027: incorporating new sources
Such as safety incidents, patients’ feedback or operational data. The goal would be to detect needs using other systems, not just explicit surveys.
2028: predictive evolution
This last phase of the project involved anticipating needs before they are explicitly stated, measuring the impact of delivered training programs, and generating proposals based on accumulated evidence.
We learned that the value lies not only in developing a technological solution, but also in gaining a deeper understanding of the problem and transforming the process while 'learning by doing'.
Another key takeaway from the LbD approach has been an organizational one: we have built the knowledge and confidence to identify where AI can genuinely add value, collaborate iteratively with a tecnology consulting partner, and retain the internal capability to continue evolving the solution over time.
AI Lead, Hospital Sant Joan de Déu
Hospital Sant Joan de Déu is one of Barcelona’s leading hospitals. Its Learning by Doing program aims to build AI awareness by enabling participants to developreal solutions for real needs while learning how to use AI along the way.
We were one of the consulting firms selected to participate in this program, and we partnered with DRIA (Direcció de Recerca, Investigació i Aprenentatge), the unit responsible for research and professional training, to help them build a solution to improve the analysis of training needs.
The challenge
The process for detecting learning needs among professionals in the hospital was extremely manual. They collected answers through a survey in which specific training actions were requested.
Cleaning, review, classification and grouping required an intensive human intervention. The resulting Excel spreadsheet required a lengthy and complex manual process that could require months.
The goal was to transform dispersed and varied answers into structured information that could be compared to support better training planning, without eliminating the expert unit’s review and analysis.
What we did
Working with the DRIA expert team, we designed and validated an MVP to automate the process, conceived as the first step in a larger architecture that could later be developed:
- Capturing the information in a structured flow
- Processing the answers with AI
- Identifying real training needs present in the answers
- Normalizing, classifying and grouping those needs
- Applying scoring and prioritization criteria
- Generating provisional results for human review
- Incorporating a human validation step before consolidating results
- Producing prioritized lists, dashboards and reports that are useful for the DRIA team
We structured this solution in five layers:
- Data collection
- Processing with AI
- Storage and traceability
- Human review
- Output generation
And, for the processing stage, we designed an AI pipeline, orchestrated with Power Automate, with three steps:
1. Normalization and classification
Initial outputs become structured information. Our system identifies the need, generates and classifies a normalized output, and determines the service and field involved.
2. Semantic grouping
Similar training needs could be expressed differently in different answers. This step of our pipeline groups them when they express the same underlying learning need, reducing duplications.
3. Scoring and prioritization
The needs that are already grouped in the previous step are evaluated using defined criteria. The outcome is a prioritized list for deciding which needs should be considered in the next training planification for professionals.
The flow works as follows:
- A file is uploaded
- A first step of the AI pipeline is executed
- An intermediate result is generated
- Human validation is requested
- If approved, the system moves forward with semantic grouping and prioritization
- The final result is stored in SharePoint, and the team is notified of the output through Teams
How to change an entire data strategy
I imagine 8wires as that sincere and honest partner that takes you out of all that noise and helps you focus on what is important, no matter how unsexy it may be, to achieve a great long-term goal.
The one who accompanies you through the hard times and helps you through the tough decisions, knowing that there is no easy road. The one who gives you the push or the tools so that you climb and be yourself the one who reaches the summits you propose in a healthy, sustainable and energetic way. And above all, the one who steps aside when he knows that he is not helping you or that he will not be able to give you what you need.
I don't know if it is helpful, but somehow I saw on the web a visual explanation of the problem in the data/technology world (I don't know if with this metaphor) before showing how it is to work with us and finally, another visual explanation of the result.
Like a bridge over troubled waters
I imagine 8wires as that sincere and honest partner that takes you out of all that noise and helps you focus on what is important, no matter how unsexy it may be, to achieve a great long-term goal.
The one who accompanies you through the hard times and helps you through the tough decisions, knowing that there is no easy road. The one who gives you the push or the tools so that you climb and be yourself the one who reaches the summits you propose in a healthy, sustainable and energetic way.
Hospital Sant Joan de Déu is one of Barcelona’s leading hospitals. Its Learning by Doing program aims to build AI awareness by enabling participants to developreal solutions for real needs while learning how to use AI along the way.
We were one of the consulting firms selected to participate in this program, and we partnered with DRIA (Direcció de Recerca, Investigació i Aprenentatge), the unit responsible for research and professional training, to help them build a solution to improve the analysis of training needs.
The challenge
The process for detecting learning needs among professionals in the hospital was extremely manual. They collected answers through a survey in which specific training actions were requested.
Cleaning, review, classification and grouping required an intensive human intervention. The resulting Excel spreadsheet required a lengthy and complex manual process that could require months.
The goal was to transform dispersed and varied answers into structured information that could be compared to support better training planning, without eliminating the expert unit’s review and analysis.
What we did
Working with the DRIA expert team, we designed and validated an MVP to automate the process, conceived as the first step in a larger architecture that could later be developed:
- Capturing the information in a structured flow
- Processing the answers with AI
- Identifying real training needs present in the answers
- Normalizing, classifying and grouping those needs
- Applying scoring and prioritization criteria
- Generating provisional results for human review
- Incorporating a human validation step before consolidating results
- Producing prioritized lists, dashboards and reports that are useful for the DRIA team
We structured this solution in five layers:
- Data collection
- Processing with AI
- Storage and traceability
- Human review
- Output generation
And, for the processing stage, we designed an AI pipeline, orchestrated with Power Automate, with three steps:
1. Normalization and classification
Initial outputs become structured information. Our system identifies the need, generates and classifies a normalized output, and determines the service and field involved.
2. Semantic grouping
Similar training needs could be expressed differently in different answers. This step of our pipeline groups them when they express the same underlying learning need, reducing duplications.
3. Scoring and prioritization
The needs that are already grouped in the previous step are evaluated using defined criteria. The outcome is a prioritized list for deciding which needs should be considered in the next training planification for professionals.
The flow works as follows:
- A file is uploaded
- A first step of the AI pipeline is executed
- An intermediate result is generated
- Human validation is requested
- If approved, the system moves forward with semantic grouping and prioritization
- The final result is stored in SharePoint, and the team is notified of the output through Teams
How to change an entire data strategy
I imagine 8wires as that sincere and honest partner that takes you out of all that noise and helps you focus on what is important, no matter how unsexy it may be, to achieve a great long-term goal.
The one who accompanies you through the hard times and helps you through the tough decisions, knowing that there is no easy road. The one who gives you the push or the tools so that you climb and be yourself the one who reaches the summits you propose in a healthy, sustainable and energetic way. And above all, the one who steps aside when he knows that he is not helping you or that he will not be able to give you what you need.
I don't know if it is helpful, but somehow I saw on the web a visual explanation of the problem in the data/technology world (I don't know if with this metaphor) before showing how it is to work with us and finally, another visual explanation of the result.
Like a bridge over troubled waters
I imagine 8wires as that sincere and honest partner that takes you out of all that noise and helps you focus on what is important, no matter how unsexy it may be, to achieve a great long-term goal.
The one who accompanies you through the hard times and helps you through the tough decisions, knowing that there is no easy road. The one who gives you the push or the tools so that you climb and be yourself the one who reaches the summits you propose in a healthy, sustainable and energetic way.
.png){kind=logo}
.png){kind=small}