In this episode of Go! For Launch, hosts Audrey, Kam, and MacKenzie sit down with Zhilin Wang, a Senior Applied Scientist at NVIDIA, to explore the cutting edge of natural language processing (NLP) and large language models (LLMs). Zhilin shares how NVIDIA’s advanced AI infrastructure powers the training of models like ChatGPT and Llama, and how his team develops techniques to make AI more helpful and human-like in conversation.Zhilin also walks us through his unconventional path into AI, from studying education and psychology at Cambridge to building an ed-tech startup that sparked his passion for NLP. He discusses his work on improving AI’s reasoning abilities, training methods, and social signal detection—and what the future holds for AI-human interaction.💡 In this episode, you’ll learn:✅ How NVIDIA’s AI chips accelerate breakthroughs in machine learning✅ What it takes to train LLMs to follow instructions and understand context✅ How psychology and education theory can shape better AI models✅ The role of internships and nontraditional learning in building an AI career✅ Whether AI chatbots will ever become indistinguishable from humans🔗 Helpful Links:🔹 Learn more about Zhilin on his GitHub page: https://zhilin123.github.io/🔹 Learn more about NVIDIA AI Research: https://www.nvidia.com/en-us/research/🔹 Explore NVIDIA’s NeMo framework for LLMs: https://developer.nvidia.com/nemo🔹 Read about Llama 2 and its impact on AI: https://huggingface.co/meta-llama🔹 Discover NVIDIA’s AI careers and internships: https://www.nvidia.com/en-us/about-nvidia/careers/

Ver todos los episodios