Castuner

Pop Goes the Stack

F5

Explore the evolving world of application delivery and security. Each episode will dive into technologies shaping the future of operations, analyze emerging trends, and discuss the impacts of innovations on the tech stack.

Słuchaj w swojej aplikacji

Odcinki

69 odcinków

  1. Vibe coding is manufacturing tech debt at machine speed 22:03
  2. AI governance: Shadow AI, real access, real risk 20:45
  3. AI broke your app? It’s probably DNS (Again) 21:39
  4. Agentic fingerprints: Detect, debug, and control AI traffic 20:25
  5. AI code quality management: Guardrails, hooks, and Open Source 22:25
  6. AI API Security: Same toolbox, new scale and new risks 24:37
  7. Stop “securing the model”—Secure the runtime instead 20:33
  8. What actually is Model Routing? A deep dive into cost, efficiency & risk 24:01
  9. AI traffic management: Load balancing vs model routing 22:41
  10. Agents go rogue: Why guardrails fail and behavior wins 21:38
  11. If your agent buys it, you bought it: Liability in AI 19:57
  12. AI blast radius: BOLA + MCP turned APIs into a 7,000-bot army 18:51
  13. Epic AI fails: Why “useful” isn’t “correct” 21:37
  14. Does your chatbot code? Why guardrails fail (and how to fix drift) 21:51
  15. The Great AI Repatriation: Why Cloud‑Only LLMs break the budget 23:22
  16. Training vs Inference: Are they the same? 21:29
  17. Round Robin is Still Dumb: Load balancing has to grow up for AI inference 21:18
  18. Mechanistic Interpretability: Debugging LLMs by reading their circuits 23:18
  19. Pop Goes the Stack will be back next week! 0:46
  20. Agent Skills: The new AI supply chain risk (and fixes) 23:13
  21. Is AI making security obsolete? 20:01
  22. Agents deleted my work: Why agents still aren’t ready for production (Yet) 23:35
  23. Agent identity: Closing the "accountability vacuum" with humans 23:16
  24. Data poisoning: You can’t patch what an LLM “learns” 23:50
  25. Local-first AI: Keep context out of the cloud 21:57
  26. DevOps meets AI agents: Risk, audit, and the Deming playbook 23:29
  27. Model routing isn’t load balancing (And that’s why you’re not ready) 20:06
  28. KV cache is the real inference bottleneck (Not GPUs) 21:09
  29. Measuring what matters: Observability for agents 20:24
  30. Alien autopsy of LLMs: Constitutions, deception, guardrails 20:54
  31. Why Prompt Filters Fail Against LLM Attacks 22:05
  32. OpenClaw: Multi-agent autonomy, secrets, and blast radius 26:34
  33. CISO Hot Takes on MCP, PQC, and Data Center Attacks 17:00
  34. AI Red Teaming in Practice: Scores, guardrails, auto-remediation 26:48
  35. Agent Identity Crisis: Access, audit, and “soul.md” 20:33
  36. VibeOps: Guardrailed agents for deterministic production 25:16
  37. WebAssembly: A programmability paradigm shift 21:36
  38. Unstructured Integration: The hidden surface area putting AI privacy & compliance at risk 24:17
  39. Logging for Giants: High-Speed Telemetry in an AI World 21:57
  40. Low-Code Automation Tools with Teeth: FlowFuse & N8N 21:35
  41. The New New User Interface: AI in your brain 18:08
  42. The Impact of Inference: Reliability 22:49
  43. The Impact of Inference: Performance 20:34
  44. The Impact of Inference: Availability 22:13
  45. Shift left into runtime: Vibe coding and AI guardrails 20:35
  46. Taking a holiday break – Pop Goes the Stack returns after New Year’s! 0:55
  47. Reshaping the web for AI agents and LLMs 22:16
  48. Five nines of wrong: Detecting drift and errors in AI systems 21:59
  49. Now you see me, now you don't: Ephemeral Auth and AI agents 24:34
  50. Holiday hiatus: Revisiting tech trends, AI, and more! 0:55