Omar Sanseviero

Career

At Google, Sanseviero worked as a software engineer and product manager, building machine learning models and infrastructure for Google Assistant and serving as a 20% product manager for the TensorFlow Graphics team. Sanseviero also spent several years at Hugging Face as Chief Llama Officer and Head of Platform and Community, where Sanseviero launched Hugging Face Spaces, built partnerships with model labs and open-source communities, set up integrations with libraries such as spaCy and Sentence Transformers, and created the developer advocacy engineering and moonshot teams.¹⁴

At Google DeepMind, Sanseviero leads a team responsible for the developer experience of AI Studio, the Gemini API and Gemma. The work spans integrations with developer tools, documentation and educational material, and the developer side of launches for models including Gemini 3, Gemma and Nano Banana.¹

Teaching and writing

Sanseviero writes about machine learning, open source and developer tools on the hackerllama blog, and co-wrote the O'Reilly book "Hands-On Generative AI with Transformers and Diffusion Models" with Pedro Cuenca, Apolinário Passos and Jonathan Whitaker. At Google, Sanseviero has co-authored developer guides for Gemma 3n in June 2025 and Gemma 4 12B in June 2026, the latter covering local inference with tools such as llama.cpp and MLX.¹⁴²³

Work and contributions

• Gemma

  Leads developer experience and launches for Gemma models, per Sanseviero's own site, and co-authored the Gemma 3n developer guide.¹²

• Gemma 4 12B

  Co-authored the Gemma 4 12B developer guide, which covers local inference with tools such as llama.cpp and MLX and points to LM Studio and Ollama for trying the model.³

• Hands-On Generative AI with Transformers and Diffusion Models Show, channel or publication

  Co-author, with Pedro Cuenca, Apolinário Passos and Jonathan Whitaker, of the O'Reilly book; the companion repository holds the book's code and exercise answers.⁴

Timeline

1. Jun 26, 2025Co-authored the Google Developers Blog guide introducing Gemma 3n.²
2. Jun 3, 2026Co-authored the Gemma 4 12B developer guide on running the model locally.³

Other documented roles

Roles the sources document that are not confirmed as current.

• Chief Llama Officer / Head of Platform and Community, Hugging Face¹
• Software Engineer and Product Manager (Google Assistant; 20% product manager for TensorFlow Graphics), Google¹⁴

Where they work

• Google DeepMind 64 people in the wiki

Related people

• Paige BaileyAlso at Google DeepMind
• Philipp SchmidAlso at Google DeepMind
• Arthur MenschAlso at Google DeepMind
• Douwe KielaAlso at Google DeepMind
• Azalia MirhoseiniAlso at Google DeepMind
• Graham NeubigLanguage models · Developer tools
• Raymond LiLanguage models · Developer tools
• Shantanu JainLanguage models · Developer tools
• Baptiste RozièreLanguage models · Developer tools
• Harrison ChaseDeveloper tools · Language models