Agentic AI and Its Guardrails:…
Guardrails written as instructions are advisory. How the enforcement boundary moved out of the prompt…
Jatin Gupta
Senior Architect @ HCL Tech
Jatin Gupta Blog
17
Controllability Over Autonomy:…
Autonomy is a means, not a goal. How to design human-in-the-loop guardrails that actually hold
18
Small vs Large: The Architecture…
They share a transformer skeleton and almost nothing else. A practical guide to SLM vs LLM architecture,…
32
Running Agentic AI on a Token…
Agent loops replay their entire context on every step. That single property is why most agentic pilots…
76
Agentic AI Automation for IT Infrastructure:…
When deterministic playbooks meet probabilistic reasoning
