LLM on Subhash Dasyam

Securing Agentic AI: Roadmap Part-10

Sun, 07 Dec 2025 20:48:00 +0400

Part 10. Implementation Roadmap

10.0 Why you need a roadmap, not a random pile of bots

You now have:

Agent patterns

Multi agent topologies

HITL designs

Threats and controls

Identity, architecture, governance

Securing Agentic AI: Enterprise Integration Part-8

Sun, 07 Dec 2025 14:51:01 +0400

8. Enterprise Integration

8.0 Why this part matters

Up to now we treated agents like a new thing. Your CISO, CIO, and Head of Architecture do not care about “new things”. They care about one question: “How does this fit into the stuff we already use to control risk?”

Securing Agentic AI: Secure Architecture Patterns Part-7

Sun, 07 Dec 2025 14:45:05 +0400

7. Secure Architecture Patterns

7.0 Why architecture beats clever prompts

Here is the uncomfortable truth: If your main security control is “We wrote a really strong system prompt”, you will lose. Not today. Maybe not this quarter. But as soon as someone finds a weird edge case or the model behaves differently after an update, your “carefully crafted” prompt will help exactly as much as a sticky note on a production firewall.

Securing Agentic AI: Identity and Access Control for Agents Part-6

Sun, 07 Dec 2025 14:40:00 +0400

6. Identity and Access Control for Agents

6.0 Why identity is the real security boundary

For classic apps, you already know the game:

User authenticates.

App runs with app identity.

Mixture of Experts (MoE): The Specialist Consultant Revolution 🏢

Sun, 13 Jul 2025 23:00:00 +0400

Building on our transformer story - if you haven’t read the complete transformer guide yet, check it out first!

Remember Our Transformer Story?

In our previous deep dive, we learned that transformers have this amazing “deep thinking step” (the Feed Forward Network) where they:

How Transformers Actually Work: The Complete Simple Guide 🤖

Tue, 08 Jul 2025 00:45:00 +0400

Ever wondered how ChatGPT, Claude, or GPT-4 actually understand and generate text? Let me break down the magic behind transformers like you’re 12 years old! 👇

Note: When I mention “117 million parameters” in examples, I’m talking about GPT-1 and BERT-base models. Modern models like GPT-4 are much, much bigger!

Graceful Degradation Strategies for GenAI Systems: Enterprise Implementation Framework

Sun, 15 Jun 2025 12:40:00 +0400

Introduction

Graceful degradation ensures systems maintain core functionality even when components fail or face performance issues, rather than experiencing complete system failure. In GenAI and inference systems, this capability becomes mission-critical as organizations increasingly rely on AI-powered applications for business operations. The approach involves systematically reducing less critical services while preserving essential operations during high-stress conditions or failures.