2026-09-25 –, Hewitt Boardroom (Capacity 35)
Agile tools like Jira excel at tracking work as it moves from “To Do” to “Done.” But when tickets also become the place where we preserve product intent, architectural decisions, and implementation context, an issue tracker unintentionally becomes a knowledge system, one that may contain all the facts without telling the full story.
This talk follows two experiments in giving AI agents the context needed to build the right product. We’ll examine what each approach made possible, where each failed, and why progressive disclosure, source authority, and maintenance cost matter as much as retrieval. The goal is not to build an all-knowing “brain,” but to help humans and agents recover the right context, understand its history, and act closer to the original intent.
Jenny is a Product Manager at Red Hat, where she drives end-to-end AI platform experiences across Red Hat OpenShift AI, spanning model serving, GPU infrastructure, and Model-as-a-Service. She focuses on making AI platforms both governable for administrators and accessible to end users. Before Red Hat, she was at Neural Magic, a Boston startup focused on GPU-efficient model inference, acquired by Red Hat in 2024. Outside of her product role, she builds with AI to reimagine how product teams operate, creating an agentic development lifecycle framework.