CSIC has its own AI

Recently, I attended the Escuela de Verano del CSIC, where I discovered that the Spanish National Research Council (CSIC) provides its own Artificial Intelligence platform, including API access for researchers and developers. This was a great opportunity to explore how these tools can be integrated into real-world AI applications.

To learn more about the platform and experiment with its capabilities, I created a GitHub repository that demonstrates how to build a simple Retrieval-Augmented Generation (RAG) system using the CSIC AI API. A RAG architecture combines information retrieval with generative AI, allowing language models to answer questions based on custom documents and knowledge sources rather than relying only on their pre-trained knowledge.

The goal of this project is to provide a practical and easy-to-follow example for anyone interested in working with the CSIC AI ecosystem. The repository includes the installation process, configuration steps, and a complete example showing how to connect documents to a language model and query them efficiently.

To make the project more accessible, I also prepared three short YouTube tutorials that explain the entire process step by step in less than three minutes.

This project allowed me to explore modern AI development practices while gaining hands-on experience with the CSIC AI infrastructure. If you are interested in Retrieval-Augmented Generation, AI APIs, or practical applications of large language models, feel free to check out the repository and the accompanying video tutorials.

https://youtube.com/shorts/sYIeb11tSgw?si=pFOYtjbdZWYpOHrz

https://youtube.com/shorts/yJz1Pv3xHoY?si=yZnTnBGTYnn4anpN

https://youtube.com/shorts/mj0ayFkE8xs?si=FbkVJLxiqWqfY77B

AIDL course finished! What's next

Last July I finally finished the "Artificial Intelligence with Deep Learning" course at the UPC (Barcelona). This has been an excellent opportunity not only to learn more about this increasing in popularity topic but also to apply it in my field. Together with three amazing colleagues and supervised by a wonderful teacher, we studied the applicability of DL for the taxonomic identification of COI sequences. My experience on this topic has been very interesting. First I really believe this project is worth of a publication so I'll work further. But it also allowed me to work with people from highly diverse fields such as Informatics and Mathematics. To converge in a way of working together, in a topic in which they were starting from scratch and I was supposed to be the expert, coming from different professional scenes... I really learned a lot and not only theory about AIDL.

Things I take home from this course: A lot of knowledge to be applied soon... stay tuned for more.

Ideas for future works.

I would really like to learn more about group managing in digital projects.

And last but not least, I knew four wonderful persons that I hope you will know them from future publications but that I won't mention right now to respect their privacy.

Scientific Paper Published: Metabarcoding identifies macroalgal composition as a driver of benthic invertebrate assemblages in restored habitats

We are excited to share the publication of a groundbreaking scientific paper co-authored by a passionate team of researchers. This work is particularly significant as it represents the culmination of the doctoral journey for the first author, Cristina Galobart, an inspiring researcher known for her deep commitment to science and ecological preservation and a good friend of mine.

In the field of ecological restoration, grasping the complex relationships between marine organisms is essential. This research looks closely at the interactions within large brown macroalgal forests and the diverse invertebrate communities that call them home.

This paper encapsulates a ten-year restoration project focused on recovering a macroalgal forest dominated by Gongolaria barbata. Through the use of advanced metabarcoding techniques, the research team examined invertebrate communities at the restored site and compared them with nearby areas.

The findings shed light on the ways these communities adapt to restoration. For example, did the invertebrate species found mirror those typically seen in healthy ecosystems? What specific roles did key species play in the restoration process? The insights offered by this research provide a glimpse into the intricate web of life that thrives beneath the ocean's surface.

If you’re intrigued by marine biology, ecological restoration, or the wonders of our oceans, stay updated with the latest research in this field. Exciting discoveries are on the horizon, and who knows what the next study will reveal.

Let's commit to keeping our oceans vibrant and thriving through dedicated research and restoration efforts!

figure 5 of the Manuscript showing Characteristic MOTUs within each assemblage type.

Link to paper:

https://www.nature.com/articles/s41598-025-93327-4