From Benchmarks to Reality: Embedding HITL in Your MLOps Stack
🎥 From the MLOps World | GenAI Summit 2025 — Virtual Session (October 6, 2025) Session Title: From Benchmarks to Reality: Embedding HITL in Your MLOps Stack Speaker: Micaela Kaplan, ML Evangelist, HumanSignal Talk Track: ML Training Lifecycle Abstract: Generative AI is reshaping how we build and deploy machine learning systems—but one thing hasn’t changed: human-in-the-loop (HITL) remains essential for quality and trust. Automated metrics can only take you so far; the real nuance often comes from human judgment about what’s right, wrong, and acceptable in context. In this session, Micaela Kaplan explores why HITL is more important than ever in the GenAI era, and how to design it into your MLOps pipelines without slowing teams down. What you’ll learn: • Practical ways to integrate HITL for monitoring, evaluation, and feedback at scale • How to engage subject-matter experts efficiently while reducing operational drag • Real-world strategies to act on the data that matters most • How to incorporate Human-in-the-Loop into your existing pipelines for trustworthy AI Whether you’re a data scientist, ML engineer, or MLOps practitioner, this talk provides a hands-on roadmap to balancing automation with the human oversight that makes GenAI work in production.
