Skip to content
View andimarafioti's full-sized avatar
507 followers · 31 following
  • Hugging Face
  • Bern, Switzerland

Achievements

Achievement: Starstruckx3Achievement: Pair Extraordinairex3Achievement: QuickdrawAchievement: Pull Sharkx3Achievement: YOLOAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Starstruckx3Achievement: Pair Extraordinairex3Achievement: QuickdrawAchievement: Pull Sharkx3Achievement: YOLOAchievement: Arctic Code Vault Contributor

Highlights

  • Pro

Organizations

@huggingface @tifgan

Block or report andimarafioti

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
andimarafioti/README.md

Hi, I’m Andi 👋

I’m an engineer working at the intersection of research and real-world systems. I build and ship multimodal AI systems, focusing on vision-language models, speech, and efficient on-device inference.

At Hugging Face, I lead multimodal research and contribute to:

  • Vision-Language Models (VLMs)
  • Speech-to-speech and conversational systems
  • Multimodal research with an emphasis on efficiency and real-world deployment
  • Robotics-facing AI systems

I enjoy building things that are both technically solid and actually usable, from research code to demos and production-ready tools.

What you’ll find here

  • Research prototypes and experimental ideas
  • Open-source tools and demos
  • Work around multimodal models, audio, and vision
  • Occasional side projects

Background

  • PhD in applied machine learning (speech and generative models)
  • Former senior ML engineer at Unity
  • Interested in small, fast, and well-engineered models

Feel free to explore, fork, or reach out.

Pinned Loading

  1. huggingface/speech-to-speech huggingface/speech-to-speech Public

    Build local voice agents with open-source models

    Python 4.9k 585

  2. huggingface/nanoVLM huggingface/nanoVLM Public

    The simplest, fastest repository for training/finetuning small-sized VLMs.

    Python 4.9k 499

  3. faster-qwen3-tts faster-qwen3-tts Public

    Real-time text-to-speech with Qwen3-TTS

    Python 1.1k 172

  4. florence2-finetuning florence2-finetuning Public

    Quick exploration into fine tuning florence 2

    Jupyter Notebook 339 31

  5. nano-parakeet nano-parakeet Public

    Pure-PyTorch Parakeet TDT inference

    Python 48 8

  6. tifgan/stftGAN tifgan/stftGAN Public

    TiFGAN: Time Frequency Generative Adversarial Networks

    Jupyter Notebook 122 13