Hugging Face is launching a new course with a two-day community event on November 15th and 16th, and the speaker lineup reads like a who’s who of modern NLP. Thomas Wolf, Hugging Face’s co-founder and Chief Science Officer, will open with a talk on transfer learning and the birth of the Transformers library. Margaret Mitchell, formerly of Google’s Ethical AI group, follows with a session on values in ML development. Jakob Uszkoreit, the Inceptive co-founder who spent over a decade at Google Brain, will present ‘It Ain’t Broke So Don’t Fix Let’s Break It’ — a title that suggests he’s more interested in challenging assumptions than polishing them.
The practical track is just as stacked. Jay Alammar, whose visual ML blog has helped millions of researchers understand everything from numpy basics to GPT-3, will give a gentle visual introduction to Transformers. Keras engineers Matthew Watson and Chen Qian will cover NLP workflows with high-level modeling APIs, and PyTorch’s Mark Saroufim will walk through model training with TorchServe. AWS is sponsoring the event by offering free compute to participants through Amazon SageMaker, which removes the usual barrier of GPU costs for anyone who wants to actually build something.
What makes this different from a typical webinar marathon is the second component: team projects. After the talks, participants will fine-tune a model on an NLP task of their choice and end with live demos. The organizers are positioning these demos as portfolio pieces for job seekers, and they’ll issue certificates of completion to anyone who successfully builds one. That’s a tangible deliverable, not just a passive viewing experience.
The Hugging Face ecosystem already reaches more than 5,000 research organizations, including Facebook AI Research, Google Research, DeepMind, Amazon Research, Apple, and the Allen Institute for AI. This event feels like an extension of that reach — bringing the same tools and people that power those labs to a broader audience of practitioners who may not have institutional compute budgets. Whether the certificate carries real weight in hiring decisions remains an open question, but the free SageMaker access alone makes the event worth a look for anyone who’s been waiting for an excuse to get hands-on with fine-tuning. Registration is open through a form linked in the announcement.
