Media Summary: GPUs get all the attention, but in inference, the real bottleneck is often memory, specifically the KV cache. In this episode of 's ... AI is no longer a lab tool—it's showing up in pipelines, production systems, and the places where “seemed like a good idea” ... Prompt injection isn't some new exotic hack. It's what happens when you throw your admin console and your users into the same ...

Pop Goes The Stack Taking - Detailed Analysis & Overview

GPUs get all the attention, but in inference, the real bottleneck is often memory, specifically the KV cache. In this episode of 's ... AI is no longer a lab tool—it's showing up in pipelines, production systems, and the places where “seemed like a good idea” ... Prompt injection isn't some new exotic hack. It's what happens when you throw your admin console and your users into the same ... Agents break the old rules of observability. Latency, throughput, and error rates still matter, but once software starts making ... Uptime used to mean reliability. But in the LLM era, five nines just means your liar is always available. Real reliability now ... The perimeter isn't where you left it. Agents are on the move, APIs are on fire, and your infrastructure is about as ready for this as a ...

Multi-model AI isn't a buzzword anymore, it's how organizations are actually operating. In this episode of 's Recorded live at in Las Vegas, this episode of Remember when were quiet little endpoints that waited politely for humans to click buttons? Yeah, that's over. Now you've ... Agents are popping up everywhere: tiny bots spinning up for a task, then dying off. They shouldn't carry long-lived credentials any ... Prompt injection has been the headline security problem for the last year, but have we been guarding the wrong layer? The 2025 API Threat Report is out, and shocker—we're still getting wrecked by injection, data leaks, and BOLA. That's Broken ...

Ops used to be a world of YAML, caffeine, and careful deploy rituals. Now it's probabilistic models, token-based cost surprises, ... Coming to you from the Hub, 's Joel Moses and guest co-pilot Oscar Spencer cut through the conference ... "It's just a chat" is the most dangerous sentence in AI. In this episode of OpenClaw is what happens when the industry looks at autonomous agents and decides they should have more autonomy, more ... AI in production isn't just another feature to ship. It's a non-deterministic system that can be socially engineered, fuzzed, and ...

Photo Gallery

Pop Goes the Stack | KV cache is the real inference bottleneck (Not GPUs) | Agentic AI
Pop Goes the Stack | DevOps meets agents: Risk, audit, and the Deming playbook | AI
Pop Goes the Stack | Crossing the streams | AI Security
Pop Goes the Stack | Measuring what matters: Observability for agents | Agentic AI
Pop Goes the Stack: Five nines of wrong - Detecting drift and errors in AI systems | LLM
Pop Goes the Stack | Taking a holiday break and returning in the New Year!
Pop Goes the Stack | The perimeter has shifted | Agentic AI
Pop Goes the Stack | Model routing isn’t load balancing (And that’s why you’re not ready) | AI
Pop Goes the Stack | CISO Hot Takes on MCP, PQC, and Data Center Attacks | Security
Pop Goes the Stack | MCP tools and AI risks: The case for slow, secure adoption | AI API
Pop Goes the Stack | Now you see me, now you don't: Ephemeral Auth and AI agents | IAM
Pop Goes the Stack: Why Prompt Filters Fail Against LLM Attacks | GenAI
View Detailed Profile
Pop Goes the Stack | KV cache is the real inference bottleneck (Not GPUs) | Agentic AI

Pop Goes the Stack | KV cache is the real inference bottleneck (Not GPUs) | Agentic AI

GPUs get all the attention, but in inference, the real bottleneck is often memory, specifically the KV cache. In this episode of #F5's ...

Pop Goes the Stack | DevOps meets agents: Risk, audit, and the Deming playbook | AI

Pop Goes the Stack | DevOps meets agents: Risk, audit, and the Deming playbook | AI

AI is no longer a lab tool—it's showing up in pipelines, production systems, and the places where “seemed like a good idea” ...

Pop Goes the Stack | Crossing the streams | AI Security

Pop Goes the Stack | Crossing the streams | AI Security

Prompt injection isn't some new exotic hack. It's what happens when you throw your admin console and your users into the same ...

Pop Goes the Stack | Measuring what matters: Observability for agents | Agentic AI

Pop Goes the Stack | Measuring what matters: Observability for agents | Agentic AI

Agents break the old rules of observability. Latency, throughput, and error rates still matter, but once software starts making ...

Pop Goes the Stack: Five nines of wrong - Detecting drift and errors in AI systems | LLM

Pop Goes the Stack: Five nines of wrong - Detecting drift and errors in AI systems | LLM

Uptime used to mean reliability. But in the LLM era, five nines just means your liar is always available. Real reliability now ...

Pop Goes the Stack | Taking a holiday break and returning in the New Year!

Pop Goes the Stack | Taking a holiday break and returning in the New Year!

Hi,

Pop Goes the Stack | The perimeter has shifted | Agentic AI

Pop Goes the Stack | The perimeter has shifted | Agentic AI

The perimeter isn't where you left it. Agents are on the move, APIs are on fire, and your infrastructure is about as ready for this as a ...

Pop Goes the Stack | Model routing isn’t load balancing (And that’s why you’re not ready) | AI

Pop Goes the Stack | Model routing isn’t load balancing (And that’s why you’re not ready) | AI

Multi-model AI isn't a buzzword anymore, it's how organizations are actually operating. In this episode of #F5's

Pop Goes the Stack | CISO Hot Takes on MCP, PQC, and Data Center Attacks | Security

Pop Goes the Stack | CISO Hot Takes on MCP, PQC, and Data Center Attacks | Security

Recorded live at #AppWorld2026 in Las Vegas, this episode of

Pop Goes the Stack | MCP tools and AI risks: The case for slow, secure adoption | AI API

Pop Goes the Stack | MCP tools and AI risks: The case for slow, secure adoption | AI API

Remember when #APIs were quiet little endpoints that waited politely for humans to click buttons? Yeah, that's over. Now you've ...

Pop Goes the Stack | Now you see me, now you don't: Ephemeral Auth and AI agents | IAM

Pop Goes the Stack | Now you see me, now you don't: Ephemeral Auth and AI agents | IAM

Agents are popping up everywhere: tiny bots spinning up for a task, then dying off. They shouldn't carry long-lived credentials any ...

Pop Goes the Stack: Why Prompt Filters Fail Against LLM Attacks | GenAI

Pop Goes the Stack: Why Prompt Filters Fail Against LLM Attacks | GenAI

Prompt injection has been the headline security problem for the last year, but have we been guarding the wrong layer?

Pop Goes the Stack | BOLA exploits: The #1 API threat and how to stop it | API Security

Pop Goes the Stack | BOLA exploits: The #1 API threat and how to stop it | API Security

The 2025 API Threat Report is out, and shocker—we're still getting wrecked by injection, data leaks, and BOLA. That's Broken ...

Pop Goes the Stack | VibeOps: Guardrailed agents for deterministic production | AIOps

Pop Goes the Stack | VibeOps: Guardrailed agents for deterministic production | AIOps

Ops used to be a world of YAML, caffeine, and careful deploy rituals. Now it's probabilistic models, token-based cost surprises, ...

Pop Goes the Stack | Agent Identity Crisis: Access, audit, and “soul.md” | Agentic AI

Pop Goes the Stack | Agent Identity Crisis: Access, audit, and “soul.md” | Agentic AI

Coming to you from the #AppWorld2026 Hub, #F5's Joel Moses and guest co-pilot Oscar Spencer cut through the conference ...

Pop Goes the Stack | Unstructured Integration: The surface area risking AI privacy & compliance | AI

Pop Goes the Stack | Unstructured Integration: The surface area risking AI privacy & compliance | AI

"It's just a chat" is the most dangerous sentence in AI. In this episode of

Pop Goes the Stack | OpenClaw: Multi-agent autonomy, secrets, and blast radius | AI

Pop Goes the Stack | OpenClaw: Multi-agent autonomy, secrets, and blast radius | AI

OpenClaw is what happens when the industry looks at autonomous agents and decides they should have more autonomy, more ...

Pop Goes the Stack | AI Red Teaming in Practice: Scores, guardrails, auto-remediation | AI

Pop Goes the Stack | AI Red Teaming in Practice: Scores, guardrails, auto-remediation | AI

AI in production isn't just another feature to ship. It's a non-deterministic system that can be socially engineered, fuzzed, and ...

Pop Goes the Stack | Securing AI Agents: Tackling the Non-Human Identity Crisis | Security

Pop Goes the Stack | Securing AI Agents: Tackling the Non-Human Identity Crisis | Security

In this episode of '