Build Your Own GPT Model Workshop by Saad Jamal

​This workshop is the 3rd in a series of building a GPT model from scratch hosted by Headstarter.

​In this session, you'll get to implement core components from the seminal paper “A Neural Probabilistic Language Model” by Yoshua Bengio, the breakthrough work that first demonstrated how neural networks can learn probability distributions over language. Participants will build a neural language model from first principles, implementing embeddings, context windows, and next-token prediction, gaining a deep understanding of the mathematical and architectural foundations that modern GPT-style models are built upon.

​This workshop will be taught by Saad Jamal, Co-Founder & CEO of Claybird (YC F25), previously MTS at Essential AI where he worked under the main author of the "Attention is All You Need" paper - Ashish Vaswani. Prior to that, Saad worked on ML systems for autonomous driving at Nuro and Tesla.