FDE, Design Engineering and Scientific Research focusing on Agents and Creative AI. Specialist in AI applications for signal processing with prior experience in audio and video computing, adaptive models and inference-based automation. Proven track record of driving innovative interdisciplinary research and high-resource deep learning deployment.

Previous participation in EvoStar, PAAMS, ICCC, ACC, VAF, ICLI, WAC. In production I have collaborated with entities Aardman Animations, Channel 4, Braga Media Arts UNESCO, Criatech, Openfield Creativelab, to name a few. Founding member of Supernova Ensemble.

Current practice: Agents and Multimodality, MAS and Integration Networks, Real-Time Inference and Cognitive Simulators.

https://luisarandas.com/pages/research
https://vecnode.github.io/

Research Projects

Collections Transformer: PDRF extending "Transforming Collections: Reimagining Art, Nation and Heritage", 1 of 5 'Discovery Projects' part of the 5-year Towards A National Collection (TaNC) programme. Funded by UKRI's Arts and Humanities Research Council (AHRC). In partnership with Tate and Decolonising Arts Institute (AH/W003341/1).

Charismatic Project: PDRF "AI-driven Storytelling for TV and Film Industries" led by Charismatic.AI, Channel 4, UAL Creative Computing Institute, Falmouth University, Aardman Animations, Sound Reactions and digital ethicist Lisa Talia Moretti, supported by InnovateUK (UKRI).

Xperimus Project: AR “Experimentation in music in Portuguese culture: History, contexts and practices in the 20th and 21st centuries” (POCI-01-0145-FEDER031380) co-funded by the European Union through the Operational Program Competitiveness and Internationalisation, in its ERDF component, and by national funds, through the Portuguese Foundation for Science and Technology (FCT/PT).

Selected Publications

Arandas, L., Pavlov, I., Grierson, M., Sutskova, O. (2026). Ollama-Based Multi-turn MAS Natural Language Conversation Prototype for Scientific Offline Testing. In: Mathieu, P., De la Prieta, F., Villani, S. (eds) Advances in Practical Applications of Agentic AI and Multi-Agent Systems, The PAAMS Collection 2026. Lecture Notes in Computer Science, vol 17174. Springer, Cham. https://doi.org/10.1007/978-3-032-41794-7_3.

Sutskova, O., & Arandas, L. (2026). Developing a GDPR Compliant Locally Deployable Experimentation Toolkit for Human to AI-Agent Interaction Testing. Cyberpsychology Section Annual Conference, The British Psychological Association (BPS), York St John University.

Arandas, L., Ionescu, I., Khan, M., Grierson, M., Carvalhais, M. (2025). All YIN No YANG: Geometric Abstraction of Oil Paintings with Trained Models, Noise and Self-reference. In: Machado, P., Johnson, C., Santos, I. (eds) Artificial Intelligence in Music, Sound, Art and Design. EvoMUSART 2025. Lecture Notes in Computer Science, vol 15611. Springer, Cham. https://doi.org/10.1007/978-3-031-90167-6_17.

Arandas, L., Sarmento, P., Grierson, M., Carvalhais, M.(2024) Editing Video Streams Through Natural Language Embeddings: Keyframe Sequencing Using Transformers. In Adriana Sá (ed) Live Interfaces journal vol. 2 (Univ. Lusófona/ CICANT). ISSN: 2975-9943.

Arandas, L., Grierson, M. & Carvalhais, M. (2023). Computing Short Films Using Language-guided Diffusion and Vocoding Through Virtual Timelines of Summaries. INSAM, Journal of Contemporary Music, Art and Technology, ISSN: 2637-1898.

Research Statement

My scientific research focuses on robust, interpretable, and multimodal Creative AI, particularly under real-world constraints such as induced hallucination, time-critical operations, and privacy requirements. 

I have developed a sustained research agenda at the intersection of:

- AI-assisted signal processing
- Multimodal agents and real-time inference
- Computing networks and system integration

With applications in audio, video and critical environments. 

My current goal is to develop reliable multimodal agents that can reason from data in a robust, interpretable, and verifiable manner. 

My work spans:

- Computer engineering advances
- System-level integration
- Deployment-oriented research

With the aim of ensuring reliability and transparency beyond controlled benchmark conditions.