Practical AI 🤖

Here's a quick (90 second) preview of the Practical AI podcast! Listen and subscribe on 👇 Apple: chlg.co/practical-ai-a… Spotify: chlg.co/practical-ai-s… Overcast: chlg.co/practical-ai-o… Android: chlg.co/practical-ai-a… RSS: chlg.co/practical-ai-r… changelog.com/practicalai/tr…

#357 We talk about self-improving AI agents, recursive learning, and the future of human-AI collaboration. 🔗 practicalai.show/357 #360 We explore securing AI agents with Zero Trust principles. 🔗 practicalai.show/360 We're excited for the second half of the year!🫶

🎊Our top episodes of Q2 are in! #351 We discuss how AI is changing the economics of software, open source, and the future of development. 🔗 practicalai.show/351 #352 We explore lessons from Claude Code Leakon AI agents, and security. 🔗 practicalai.show/352

Watch the video version here: youtube.com/watch?v=Pd1T7F… Listen to the audio version here: practicalai.show/364

In this episode, @CoreySandersWA , SVP of Product at @CoreWeave, joins us to discuss AI-native infrastructure, GPUs, training vs. inference, agentic development, and why the future of software will be built around AI-first experiences. 🎥 Watch or 🎧 Listen: [Link below]

Watch the video version here: youtube.com/watch?v=1PtFge… Listen to the audio version here: practicalai.show/363

What does it take to move #AI agents from demos to production? Hamza Tahir from ZenML explains how MLOps principles are shaping generative AI, from workflows and agent harnesses to fleets, durable infrastructure, and ZenML's new project, Kitaru. Watch/listen here👇 [Link below]

Across all three conversations, one message stood out: the future of AI isn't just about building more capable models, it's about building systems that are reliable, secure, and worthy of trust. Which topic sparked your thinking the most this month? 👇

3️⃣ Building Trust in AI Agents with AIUC-1 Emil Lassen joined us to discuss how standards, certification, audits, red teaming, and insurance can work together to create trustworthy AI systems and accelerate enterprise adoption. 🔗Listen: practicalai.show/361

June Recap: Three episodes. One recurring theme: building AI we can actually trust. Here's what we explored in June on the podcast. If you missed any of these episodes, now's a great time to catch up. 🧵👇

Watch the video version here: youtube.com/watch?v=KODaNT… Listen to the audio version here: practicalai.show/362

AI image generation has come a long way. This week, @dustin_podell from @bfl_ai joins us to discuss: ➡️Diffusion vs. flow matching ➡️How modern image models work ➡️The FLUX family of models ➡️Running image generation locally Watch/listen here: [Link below👇]

Throwing it back to last year’s @midwestaisummit! 🧠 Our hosts took the stage to talk practical AI, and the energy was unmatched. Missed it last time? Don’t miss this year. 🗓️ Oct. 15, 2026 📍 Indianapolis, IN 🎟️ Tickets: [Link below👇] #MidwestAISummit #PracticalAI #AI

Listen here: practicalai.show/361

📋How do we build trust in #AI agents before the AI hailstorm arrives? Emil Lassen from @aiunderwriting joins the show to discuss: • AIUC-1 certification • AI standards, audits & insurance • Red teaming AI agents • Enterprise AI trust & security 🎧 Listen now: [link below]

Safety challenges emerge when AI moves from generating text to taking actions. And as agentic systems become ever more capable, the competitive advantage they will offer early adopters is staggering in its imbalance.

The rise of complex agentic systems has necessitated: → Rapid maturity in harnesses, governance, workflows, integration, safety controls → Application of zero-trust principles → Allocation of infrastructure → Shift from model selection to agentic system design

🚨One of the major themes addressed on Practical AI during the past year is the focal transition from large language models (LLM) to agentic systems in which the LLM orchestrates effects upon the digital or real world. 🧵👇