0:00-46:39
Streams straight from the publisher. podnod never proxies or re-hosts episode audio.
What does it take to move AI agents from demos to reliable production systems? In this episode, Hamza Tahir explores how MLOps principles are shaping the future of generative AI, covering workflows, agent harnesses, fleets, and the infrastructure needed to build durable, scalable systems. The conversation dives into open source tools, production challenges, and how ZenML's new project, Kitaru, helps developers build resilient, replayable, and observable agent systems.
Featuring:
Links:
Sponsors:
- Framer: The enterprise-grade website builder that lets your team ship faster. Get 30% off at framer.com/practicalai
Upcoming Events:
- Register for upcoming webinars here!
- Midwest AI Summit 2026
LinkedIn
linkedin.comWebsite
datadan.ioGitHub
github.comX
x.comZenML
zenml.ioKitaru
zenml.ioMachine Learning Tools Landscape v2 (+84 new tools)
huyenchip.comframer.com/practicalai
framer.comupcoming webinars here
practicalai.fmMidwest AI Summit 2026
midwestaisummit.com
- 0:00Welcome to Practical AI!
- 0:41Introducing the Guest
- 2:48The Evolution of MLOps and Agent Technology
- 4:35Understanding Workflows in AI and Agents
- 8:43Challenges of Durability in AI Agents
- 10:59The Future of Agents in Cloud Environments
- 13:00The Role of Harnesses in Agent Development
- 16:13The Tension Between Open and Proprietary Harnesses
- 20:45Sponsor: Framer
- 21:49Understanding Agent Architectures
- 25:15Building Durable Agent Systems
- 29:48Challenges in Agent Development
- 34:20The Future of Agent Infrastructure
- 39:03Optimizing Agent Performance
- 42:00Looking Ahead: Opportunities and Challenges
- 45:58Outro