Operations | Monitoring | ITSM | DevOps | Cloud

AI has emerged as a defining force in modern technology, spearheading transformation across industries. Yet, despite its promise to revolutionize workflows and unlock unprecedented efficiency, most DevOps organizations face significant hurdles in adopting AI safely and effectively. Concerns about complexity, scalability, and governance hold many decision makers back.

I ran a mini-red-team exercise against Azure OpenAI, testing common jailbreak tactics and prompt engineering attacks. Microsoft Defender for Cloud caught nearly everything. The key? A feature called User Prompt Evidence that turned vague alerts into precise, real-time context. If you’re running AI workloads in Azure and haven’t turned this on yet, you’re playing defense blindfolded.

Most IT leaders already understand that AI is the future of operations, but too many are still treating it like it’s so far off. The irony? Waiting is exactly what’s costing them the most. While businesses obsess over budget cuts, resource constraints, and service quality, one truth remains: delays in adopting AI for IT operations are compounding operational inefficiencies, inflating labor costs, and stalling digital progress. AI isn’t just about innovation; it’s about scale.

On a recent episode of The Product Manager podcast, Cortex CEO Anish Dhar joined host Hannah Clark to challenge a growing narrative: that software engineering is obsolete in the age of AI. His take? Engineering isn’t disappearing, it’s maturing. At Cortex, we work with some of the most forward-thinking engineering organizations at companies like Canva and Fanatics.

AI is changing how we build software — but debugging code still comes down to having context. One minute the model’s performance is cruising. The next, you’re hit with a KeyError from a tool you forgot existed, triggered by a model that silently timed out, and a retrieval call that returns... nothing, or 11 “Let me try this a different way" messages before failure. You’re stitching together LLM calls, agents, vector stores, and custom logic. Then hoping it holds up in prod.