Overview
In this episode of The Dashboard Effect, Brick Thompson and Landon Oaks take on the fear that has been circulating in software engineering communities: that AI tools are coming for developer jobs. Their argument is more nuanced and more optimistic than either the doom narrative or the uncritical enthusiasm that tends to dominate this conversation. AI is changing what great engineering looks like, but the demand for skilled engineers is not going away. What is going away is the ceiling on what a skilled engineer can accomplish.
The conversation is grounded in the real experience of a team that has been using these tools in production data engineering work and has watched the capabilities evolve significantly over a short period of time. See how Blue Margin’s Managed Data Service applies the latest AI engineering tools to deliver faster, more efficient data solutions for the organizations it serves, with the human expertise and architectural judgment that keeps quality, security, and context intact.
What This Episode Covers
Overcoming Resistance to AI Tools (0:00 – 2:57)
Many developers formed their opinions about AI coding tools based on early experiences with models that were genuinely unreliable, and those opinions have not been updated to reflect how dramatically the tools have improved. The hosts make the case for re-evaluation with an open mind, arguing that the gap between where the tools were in late 2025 and where they are now is large enough to invalidate conclusions drawn from the earlier experience.
The New Engineering Mindset (3:00 – 6:59)
The shift the hosts describe is from writing code to orchestrating it. Great engineers are increasingly using AI to handle repetitive, boilerplate work while applying their own architectural expertise and judgment to review, refine, and ensure the output is modular, efficient, and maintainable. The engineering value moves up the stack toward design and taste rather than disappearing, and the engineers who make that transition are significantly more productive than those who do not.
Building an AI Enablement Culture (7:00 – 9:14)
Leaders who want their teams to adopt AI tools effectively should invest in creating an environment where early adopters share successful workflows and demonstrate what is possible to colleagues who are more skeptical or less experienced with the tools. That peer-to-peer knowledge sharing is faster and more credible than top-down mandates, and it produces the kind of organic adoption that turns individual productivity gains into team-wide capability improvements.
The Future of Software Engineering Jobs (9:15 – 12:29)
The hosts push back on the apocalyptic framing of AI’s impact on software careers. Demand for software professionals continues to grow, and the engineers who will thrive are the ones who integrate AI into their workflows rather than resisting it. The skill set is shifting, but the profession is not disappearing. What is changing is the floor of what a competent engineer can produce and the ceiling of what an exceptional one can accomplish.
Tooling Advancements: Claude Code and ChatGPT Codex (12:30 – 14:09)
Recent updates to tools like Claude Code and ChatGPT Codex have significantly changed what AI-assisted engineering looks like in practice. The capabilities are genuinely impressive, but the hosts are clear about the ongoing requirement for human involvement. Code quality, security, and contextual judgment still require a person in the loop, and engineers who treat these tools as autonomous rather than assistive tend to produce outputs that look functional but carry hidden problems.
Who It’s For
This episode is worth your time if you are a software engineer or data developer who formed an opinion about AI coding tools based on early experiences and has not revisited those tools recently, a team lead or engineering manager trying to understand how to build a culture that embraces AI tooling without sacrificing the code quality and architectural standards the team is responsible for, a business or technology leader evaluating how AI is changing the economics and capabilities of software development teams, or anyone trying to form a grounded view of what AI means for the engineering profession beyond the extremes of replacement and irrelevance.
Why It’s Worth a Listen
The resistance narrative around AI coding tools is understandable but increasingly outdated, and this episode makes the case for updating it with enough specificity to be convincing. The hosts are not cheerleaders for the technology. They are practitioners who have watched it improve in real time and are describing what that improvement actually means for how engineers work, not what it might mean in a hypothetical future.
The orchestration mindset shift is the most important reframe in the episode. Engineers who understand their value as residing in architectural judgment, code review, and contextual expertise rather than in raw coding output are positioned to become significantly more productive with AI tools rather than threatened by them. That reframe is not just reassuring. It is actionable, and it changes how engineers should be thinking about their own skill development going forward.
And the tooling update is worth hearing for anyone who wrote off AI coding assistants based on experiences from even six months ago. The pace of improvement in this space is fast enough that conclusions drawn from earlier models are genuinely unreliable guides to what the current tools can do, and this episode provides a current and credible benchmark for reassessing that judgment.
