Blackstone’s Chief Technology Officer, John Stecher, has shared his perspective on how the entry-level software engineering landscape has transformed since he graduated in 2001. Speaking with Business Insider, Stecher acknowledged that the field looks “totally different” now compared to when he began his career. Back then, computer science was seen as an “ultra-nerdy” path, often attracting highly introverted individuals like himself. He recalled spending countless hours poring over textbooks to learn how to write code—a process that required deep, self-directed study. Today, he notes, engineers can simply search online or use AI tools like Claude to get answers instantly. Stecher also pointed to the dramatic growth in computer science enrollment at universities. At the University of Wisconsin-Madison, where he earned his degree, the number of students in the program has surged from about 260 in 2001 to roughly 2,500 expected in fall 2025. Despite the shift in how people learn, Stecher said the core technical competencies required for success in engineering haven’t changed. What has changed, he noted, is the level of skill many new engineers bring to the table. “I’m blown away every year,” he said. “There is no way in hell I would have gotten a job with the skills that some of these people have today.” However, he also highlighted a growing challenge: fewer opportunities for hands-on learning and mentorship. In the past, entry-level engineers were often given real-world tasks—like writing functional tests—that helped them understand and own parts of a larger system. Now, with AI-powered tools automating many of these routine coding tasks, new engineers are missing out on the kind of foundational experience that once helped them grow. “AI tooling can do that for you,” Stecher said, “but that means we have to find new ways to teach the next generation how to think like engineers.” To address this, Blackstone has implemented a “buddy system” that pairs junior engineers with more experienced colleagues. This mentorship model helps early-career hires learn not just how to use AI tools, but also when to question their output—understanding when the tool is correct and when it’s not. Stecher emphasized that the most valuable skill today isn’t just coding ability, but the capacity to break down complex problems, apply the right tools, and critically evaluate their results. “That is a very important skill set that people need to pick up and work with,” he said.