Bridging the Skills Gap: Insights from Agentic Coding Training
Daniel Jones and Benedict Stemmelt dive into the technical and organizational shifts of agentic software engineering. They move past the hype to discuss the transition from manual coding to building software factories that automate feature delivery. The conversation explores critical technical hurdles, including context window reasoning degradation and the necessity of reverse engineering model defaults to steer agents effectively. By referencing DORA 2025 metrics, they illustrate why AI adoption accelerates high-performing teams while exposing systemic bottlenecks in lower-maturity organizations.
Episode Transcript
Daniel Jones (00:02) Welcome to the Waves of Innovation podcast. am DJ, your host. In this episode, I am talking to Benedict Stemmelt from Hackers and Wizards who help people with agentic coding. They deliver training to organizations to upskill their developers and help with the change that's neede...
Episode Highlights
- •Senior leaders reclaim the joy of building by using agentic tools to bypass environmental setup friction.
- •Effective training must prioritize the unhappy path to teach developers how and why models actually fail.
- •Reasoning drops sharply after 30,000 tokens, requiring engineers to prioritize aggressive context curation for accuracy.
- •Developers must reverse engineer model defaults to effectively steer agentic outputs toward project-specific architectural requirements.
- •The rise of Software Factories shifts engineering focus from manual coding to designing autonomous production machines.
- •AI acts as a multiplier, accelerating high-maturity teams while slowing down organizations with existing systemic bottlenecks.
- •Aligning on coding standards is a prerequisite for successful department-wide adoption of agentic engineering toolsets.
