Christin Pohl
AI Infrastructure Engineer at Microsoft focused on LLM inference and GPU optimization.
Christin Pohl works on AI infrastructure at Microsoft, where she helps enterprises figure out the logistics of large language models. Her day-to-day involves choosing the right GPUs, running full-cycle LLM training, and dialing in cost per token.
She started building chatbots in 2018. Today, she breaks down the trade-offs of inference optimization, teaching engineering teams when to rely on managed APIs and how to handle self-hosted models using PyTorch-based frameworks.
Before moving to Switzerland, she studied computer science across Berlin, Toronto, and Harvard.
