Agentic AI
The Agentic Development Transformation
Adapt faster with agentic development: redefine technical debt, accelerate delivery, and tackle challenges in months, not years.
Key takeaways: Agentic development is transforming software engineering by prioritizing design, rapid iteration and adaptability. Dell Technologies enables faster delivery and rethinking technical debt with innovative approaches like the Dell Data Orchestration Engine.
I’ve led engineering teams through multiple transformations over the past two decades, from waterfall to agile, from 18-month release cycles down to three or four months. But what’s happening right now with agentic development is easily the largest transformation I’ve ever experienced.
Even in just the last three months, the approach to software development has completely changed.
Here’s what’s different: when you think about the amount of time it takes to code something, it compresses significantly. The upfront work becomes much more important. The major point is we are spending effort up front to define the context to enable rapid development. This requires prioritizing design and system thinking. The actual building of the software is super fast. But validation of what was built becomes much more critical.
This changes everything about how we think about what’s possible.
We’re now having conversations internally about decisions that were made 10 years ago. We have parts of our software that have been in place for a decade; they’re proven, but they’re very slow to iterate on. Now we’re asking: do we just throw that away and rebuild it? Because we can rebuild it so quickly.
That’s a fundamentally different way of thinking about technical debt and refactoring.
My perspective on this transformation is that flexibility is key. If anyone says they have the answer right now, they don’t. There are a bunch of things that are going to be tried; some will have good outcomes, others won’t. If you’ve set up a rigid environment where you don’t have the capability to adapt, you’re going to fail.
We’re operating on hypotheses. We believe certain things are correct, and we’re testing them. Some we’ve already tested and learned from. Others we’re testing now. But ultimately, we’re trying to be an adaptable organization that can understand where the industry’s going.
What customers thought was most important six months ago isn’t what matters now. Six months ago, we thought AI development was about code completion and generative AI. Now that feels like ancient history. Agentic development is completely different.
We’re starting to have conversations with customers about this rapid delivery capability, and there’s a natural tension. Customers are cautious about code adoption. They have checks and balances in place. But if they want to stay ahead, they’re going to have to consume software more rapidly. That goes against many of the processes they have in place now.
That push and pull is going to be a major discussion point going forward.
The biggest change for the storage market broadly is the ability to think about doing things you otherwise wouldn’t have been able to do because of the time they would take. It opens up our ability to rethink our approach and take on larger challenges. Things that might have been a two-year effort with questionable ROI can now be done in two months. The whole dynamic changes.
That’s the world we’re in, and the organizations that can adapt fastest will lead it.
Read on!
