cv

Basics

Name Dario Cioni
Label ELLIS PhD Student, Queen Mary University of London
Email d.cioni@qmul.ac.uk
Url https://ciodar.github.io
Summary ELLIS PhD student at Queen Mary University of London, in collaboration with Tavus, working on generative models and multimodal learning for affective, embodied conversational agents. Previously Senior AI Engineer at the PwC Italy AI Center of Excellence, leading production-scale video and audio generation systems.

Education

  • 2026.05 - Present

    London, UK

    Ph.D. in Computer Science (ELLIS PhD Program)
    Queen Mary University of London
    Multimodal representations for affective embodied conversational agents
    • Supervisors: Prof. Ioannis Patras, Prof. Georgios Tzimiropoulos
    • ELLIS co-advisor: Prof. Nicu Sebe, University of Trento
    • In collaboration with Tavus
  • 2021.09 - 2024.04

    Florence, Italy

    M.Sc. in Artificial Intelligence
    University of Florence
    Thesis: Forensic Techniques for Synthetic Image Detection and Attribution
    • 110/110 with Honours
    • Supervisors: Prof. Lorenzo Seidenari, Prof. Andrew D. Bagdanov; co-supervisors: Prof. Ioannis Patras, Dr. Christos Tzelepis
  • 2015.09 - 2021.04

    Florence, Italy

    B.Sc. in Computer Engineering
    University of Florence
    Thesis: Convolutional Neural Networks for Object Counting in Thermal Imagery
    • 110/110
    • Supervisor: Prof. Andrew D. Bagdanov

Work

  • 2024.07 - 2026.04

    Milan, Italy

    Senior AI Engineer
    PwC Italy - AI Center of Excellence
    Research and development of generative and multimodal AI systems; AI education programs.
    • Led a team of 7 building a video generation suite (text-to-video, image-to-video, any-to-audio) on open-source DiT models, self-hosted on an internal inference platform; optimized inference with post-training quantization and step distillation.
    • Designed and built multi-agent (agentic AI) systems for internal products and client-facing applications.
    • Developed retrieval-augmented generation pipelines and an internal LLM evaluation pipeline.
    • Co-designed and taught AI training modules for PwC Italy and university partners, including hands-on labs for executives and graduate students.
  • 2023.10 - 2024.02

    London, UK

    Visiting Research Assistant (Master's Thesis)
    Queen Mary University of London - Centre for Multimodal AI
    Research on forensic techniques for synthetic image detection and origin attribution.
    • Used general-purpose foundation model features (CLIP, DINOv2) for open-set synthetic image origin attribution, reaching state-of-the-art results on diffusion-generated images.
    • Led the project end to end: ideation, literature review, experiments and paper writing. Best Workshop Paper Award at TWYN @ ECCV 2024.
  • 2023.02 - 2023.07

    Florence, Italy

    Research Assistant
    University of Florence - Media Integration and Communication Center
    Vision-language research on diffusion-based augmentation for cultural heritage datasets.
    • Built a diffusion-based multimodal augmentation pipeline improving image captioning and multimodal retrieval (ICCVW 2023).
  • 2015.08 - 2021.09

    Florence, Italy

    Product Owner & Software Developer
    Hermes Trade S.r.l.
    ERP/CRM development across the full software development lifecycle, alongside Bachelor's studies.
    • Designed and led the development of an ERP/CRM application as backend developer and data engineer.

Publications

Teaching

Awards

Projects

  • Deep Compression
    • Reproduced the three-step compression pipeline on an ImageNet subset and MNIST.
  • Deep Learning Applications
    • Built end-to-end implementations of research papers for coursework and experimentation.

Skills

Research areas
Diffusion and flow-matching models
Audio-visual generation
Vision-language models
Affective computing
Synthetic media forensics
Engineering
PyTorch
Distributed training
Quantization and distillation
Model serving
Agentic AI systems

Languages

Italian
Native speaker
English
Full professional (IELTS Academic 8.0)

Interests

Research interests
Generative Models
Multimodal Learning
Affective Computing
Conversational Agents
Computer Vision