ServiceNow AI Research

Miguel Muñoz-Mármol

Miguel Muñoz-Mármol

AI Developer

Adaptive Agent

Miguel is a Staff AI Research Developer at ServiceNow AI Research, where he works on the evaluation and security of agentic AI systems. His research interests span AI agent security evaluation, prompt injection and adversarial robustness, guardrails and defense mechanisms, enterprise AI benchmarking, and evaluation methodology, with a consistent focus on grounding measurement in realistic production conditions.

He has broad expertise in the research, development, and production of machine learning systems, with a background spanning nearly two decades of software engineering and over a decade in AI and NLP. Before joining ServiceNow, he built production multilingual NLP systems at Maluuba (acquired by Microsoft), led machine learning teams at Adlib Software, and conducted research in biomedical AI at Universidad Politécnica de Madrid.

Miguel holds a BSc in Computer Science and Software Engineering and an MSc in Artificial Intelligence from Universidad Politécnica de Madrid.

Interests
  • Natural Language Processing
  • Software Engineering

Publications

DRBench: A Realistic Benchmark for Enterprise Deep Research. Amirhossein Abaskohi, Tianyi Chen, Miguel Muñoz-Mármol, Curtis Fox, Amrutha Ramesh, Étienne Marcotte, Xing Han Lu, Nicolas Chapados, Spandana Gella, Christopher Pal, Alexandre Drouin, Issam H. Laradji. At International Conference on Learning Representations,  2026.

Paper Cite Code

How to Train Your LLM Web Agent: A Statistical Diagnosis. Dheeraj Vattikonda, Santhoshi Ravichandran, Emiliano Penaloza, Hadi Nekoei, Thibault Le Sellier De Chezelles, Megh Thakkar, Nicolas Gontier, Miguel Muñoz-Mármol, Sahar Omidi Shayegan, Stefania Raimondo, Xue Steve Liu, Alexandre Drouin, Alexandre Piche, Alexandre Lacoste, Massimo Caccia. At Workshop at the Neural Information Processing Systems (NeurIPS),  2025.

Paper Cite

How to Train Your LLM Web Agent: A Statistical Diagnosis. Dheeraj Vattikonda, Santhoshi Ravichandran, Emiliano Penaloza, Hadi Nekoei, Thibault Le Sellier De Chezelles, Megh Thakkar, Nicolas Gontier, Miguel Muñoz-Mármol, Sahar Omidi Shayegan, Stefania Raimondo, Xue Steve Liu, Alexandre Drouin, Alexandre Piche, Alexandre Lacoste, Massimo Caccia. At Neural Information Processing Systems (NeurIPS),  2025.

Paper Cite

How to Train Your LLM Web Agent: A Statistical Diagnosis (Oral). Dheeraj Vattikonda, Santhoshi Ravichandran, Emiliano Penaloza, Hadi Nekoei, Thibault Le Sellier De Chezelles, Megh Thakkar, Nicolas Gontier, Miguel Muñoz-Mármol, Sahar Omidi Shayegan, Stefania Raimondo, Xue Steve Liu, Alexandre Drouin, Alexandre Piche, Alexandre Lacoste, Massimo Caccia. At Workshop at the International Conference of Machine Learning (ICML),  2025.

Paper Cite