Microsoft is integrating AI into its development cycle, which is expected to fix hundreds of security holes in its products. The company’s approach to using AI in software development has been explained by Charles Lamanna, Executive Vice President of Copilot, Agents and Platform.
Lamanna has emphasized that his team’s biggest milestone this year is changing how they build software, thanks to agentic coding and other AI technologies.
Microsoft’s New Approach to Software Development
Engineers are no longer writing each line of code by hand, but are instead directing agentic systems to do the work for them. This allows them to focus on more important decisions like strategy, architecture, and quality.
Products are not constrained by coding anymore, the bottleneck has shifted to making the right decisions that will result in the right product. Lamanna says that even though AI has enabled engineers to build lots of more features, he is guiding his team to build the same number of features but use that surplus time to focus on reliability, performance, and security.
Improvements and Challenges
A notable improvement as a result of this effort is that Power Automate tickets have dropped by double-digit percentages year-over-year. Lamanna has emphasized that customer success metrics and KPIs should stay the same, regardless of AI.
Related: Software Predictions for Businesses in 2026
However, Lamanna has admitted that Microsoft obviously hasn’t figured out everything, and the firm is struggling to tackle security bugs faster than AI can discover them.
It’s likely that as Microsoft continues to integrate AI into its development cycle, the company will face new challenges and opportunities. By focusing on the fundamentals of software development and using AI to improve processes, Microsoft may be able to stay ahead of the curve in the rapidly evolving tech industry.
Adopting AI in Software Development
Lamanna notes that engineering teams shouldn’t bolt AI into their processes just for the sake of it, but fundamentally reshape how they work around the technology itself. Successful teams should highlight the importance of adopting AI so that others are encouraged to do so too.
This approach has led to Microsoft itself shipping products from a concept stage to mass market in a matter of months, something that was unfathomable a few years ago. As the company continues to push the boundaries of what’s possible with AI in software development, it will be interesting to see how the tech industry as a whole responds to these changes.
