Podcast · 157 episodes
Practical AI
by Practical AI LLC · English
Making artificial intelligence practical, productive & accessible to everyone. Practical AI is a show in which technology professionals, business people, students, enthusiasts, and expert guests engage in lively discussions about Artificial Intelligence and related topics (Machine Learning, Deep Learning, Neural Networks, GANs, MLOps, AIOps, LLMs & more). The focus is on productive implementations and real-world scenarios that are accessible to everyone. If you want to keep up with the latest advances in AI, while keeping one foot in the real world, then this is the show for you!
Latest episodes
From AGENTS.md to Enterprise DeploymentDeploying agent workflows in enterprise environments using “agents.md” and a platform approach (VMware Tanzu/Cloud Foundry), including MCP tool access, identity/SSO, memory services, and an AI…24 Sep 2026 · 49 min
How to get discovered in AI searchHow brands get “discovered” in AI search/LLM answer engines (AEO/GEO/AEO), and how to measure and optimize visibility beyond traditional SEO.17 Sep 2026 · 55 min
Computer-Use Agents and the Future of the Agentic InternetComputer-use agents enabling an “agentic internet,” plus the role of MCP/agentic interfaces, security and trust concerns, and how “harnesses” (tooling/system prompts) blur into model behavior.10 Sep 2026 · 56 min
All episodes
From AGENTS.md to Enterprise DeploymentDeploying agent workflows in enterprise environments using “agents.md” and a platform approach (VMware Tanzu/Cloud Foundry), including MCP tool access, identity/SSO, memory services, and an AI…24 Sep 2026 · 49 min · 20 chapters
How to get discovered in AI searchHow brands get “discovered” in AI search/LLM answer engines (AEO/GEO/AEO), and how to measure and optimize visibility beyond traditional SEO.17 Sep 2026 · 55 min · 18 chapters
Computer-Use Agents and the Future of the Agentic InternetComputer-use agents enabling an “agentic internet,” plus the role of MCP/agentic interfaces, security and trust concerns, and how “harnesses” (tooling/system prompts) blur into model behavior.10 Sep 2026 · 56 min · 20 chapters
Less about Models; More about ArchitectureThe episode argues that enterprises should stop focusing on “models” and instead design full AI architectures that are sovereign, safe, governed, and operationalized.3 Sep 2026 · 46 min · 12 chapters
Building the Foundation for the Agentic AI EraBuilding the foundation for “agentic AI” by creating neutral standards and protocols so agents can interoperate across tools, companies, and regions; also how to drive enterprise adoption via change…28 Aug 2026 · 45 min · 13 chapters
AI Proficiency: From Users to Builders“AI proficiency” as a workforce transformation ladder from non-users to “non-technical builders,” plus how to scale agentic AI safely using an AI control plane.25 Aug 2026 · 56 min · 20 chapters
Models, Harnesses, and Multi-Agent SystemsEducational, no-guest episode defining AI models, open-weight vs closed models, and how agents and multi-agent “harnesses” enable autonomous, goal-driven work (including fleets/swarm concepts) and…6 Aug 2026 · 50 min · 17 chapters
Reconstructing how OpenAI agents attacked Hugging FaceHow OpenAI agentic testing led to a sandbox escape, then a multi-step compromise of Hugging Face, including attempts to steal private benchmark answers and credentials; also how Hugging Face…30 Jul 2026 · 44 min · 14 chapters
Surviving the New Economics of a Post-Agentic WorldThe “new economics” of a post-agentic world—how long-lived, multi-agent systems and agent-to-agent workflows change business budgets, enterprise software value, and labor roles; includes IBM’s stock…23 Jul 2026 · 36 min · 14 chapters
The Future of AI Infrastructure with CoreWeaveHow CoreWeave is building AI-specific infrastructure and developer workflows for training and inference, enabling “AI loop” iteration with observability, experiment tracking, and agent-assisted…17 Jul 2026 · 50 min · 20 chapters
Building Durable AI AgentsBuilding durable, reliable AI agents for enterprise/cloud use. Hamza argues that classic ML pipeline principles (graphs, productionization, retryability) still apply, but agents add cycles, real-time…9 Jul 2026 · 47 min · 22 chapters
Image Generation and Visual Intelligence with Black Forest LabsState of image/video/audio generation and “visual intelligence,” comparing diffusion vs autoregression, explaining Black Forest Labs’ flow-matching approach, and how context-aware editing enables…2 Jul 2026 · 48 min · 18 chapters
AIUC-1: Building trust in AI agentsBuilding trust in agentic AI via AIUC-1 standards, third-party audits, and insurance; how enterprises can operationalize AI safety/security rather than treating frameworks as aspirational.25 Jun 2026 · 45 min · 15 chapters
Zero Trust for AI AgentsZero Trust for AI agents, based on an Anthropic framework for deploying autonomous agents in enterprises.11 Jun 2026 · 47 min · 16 chapters
Breaking down the 2026 Stanford AI Index ReportHosts break down the 2026 Stanford AI Index Report, focusing on capability trends, geopolitics, benchmarks, robotics, responsible AI, investment/talent, and education/productivity impacts.4 Jun 2026 · 47 min · 19 chapters
Rebooting Enterprise AI with MCP and KubernetesRebooting enterprise AI by using Model Context Protocol (MCP) as a controlled “tooling membrane” and deploying MCP servers with Kubernetes, plus identity, governance, and proxying for secure agentic…28 May 2026 · 48 min · 14 chapters
Hermes Agent: Agents that grow with youNoose Research’s evolution and the Hermes Agent project—open-source “agents that grow with you.” The episode argues that agent value comes from the harness (tools, memory, skills, state) plus the…21 May 2026 · 52 min · 17 chapters
U.S. Congressman Beyer on AI challenges facing America and the WorldU.S. AI policy and safety amid rapid model progress; cybersecurity and surveillance risks; regulation approach (federal vs state vs international); AI-driven job displacement and economic adaptation;…14 May 2026 · 45 min · 20 chapters
The Myth of Model Wars: Open vs Closed AI in 2026The episode argues that “open vs closed” model performance is no longer the main story for 2026.7 May 2026 · 42 min · 14 chapters
The mythos of Mythos and Allbirds takes flight to the neocloudAllbirds’ pivot from shoes to AI compute infrastructure, the rise of “NeoCloud” (AI-native cloud for GPU-first workloads), Anthropic’s rumored “Mythos” frontier model used in a gated security…23 Apr 2026 · 45 min · 19 chapters
Open Source Self-Driving with Comma AIComma AI’s OpenPilot open-source self-driving/ADAS stack and how it delivers highway autonomy via an add-on device, trained with simulation and a “world model,” plus what’s needed for broader…16 Apr 2026 · 46 min · 22 chapters
Post-Mortem of Anthropic's Claude Code LeakPost-mortem of the April 1, 2026 leak of Anthropic’s Claude Code, including a supply-chain attack and what the leak reveals about agent “harness” IP and secure agent design.9 Apr 2026 · 45 min · 19 chapters
Agentic Coding and the Economics of Open SourceHow “vibe coding”/agentic coding changes incentives and attention in open source, potentially reducing human visibility and harming open-source sustainability.2 Apr 2026 · 49 min · 22 chapters
AI at the Edge is a different operating environmentEdge AI (“AI at the edge”) in 2026: what “edge” means (not in the cloud), why it’s constrained (size/power/connectivity/cost/reliability/latency/privacy), and how modern AI architectures use smaller…25 Mar 2026 · 47 min · 17 chapters