More Hanselminutes with Scott Hanselman episodes
Published 19 Feb 2026
Duration: 00:41:24
Software development is advancing with AI-driven code generation and productivity tools, but human judgment and quality assurance are still essential, as AI's creativity and originality limitations mean it shouldn't replace human developers entirely.
What does it take to design a programming language from scratch when the target isnt just CPUs, but GPUs, accelerators, and the entire AI stack? In th...
The podcast examines how AI is reshaping software development, influencing areas such as code generation, productivity, and the overall developer experience. While AI tools excel at pattern recognition and replication, they are noted to lack true creativity and originality, often producing standard, unoriginal results. An example given is an AI-generated C compiler, which demonstrated strong language translation skills but failed to deliver innovative outcomes. The discussion reflects on past developer resistance to new technologies, like the shift from Objective-C to Swift, and draws comparisons to current hesitations around AI integration.
There is an emphasis on the continued need for human judgment and quality assurance in software engineering, even as AI becomes more prevalent. Additionally, the conversation highlights the growing demand for advanced programming tools that can effectively leverage modern hardware such as GPUs and AI accelerators. A proposed solution is a new programming language called Mojo, designed to offer improved performance, flexibility, and ease of use across different hardware platforms. The overall message stresses the importance of balancing AI advancements with human expertise and the development of tools that support the evolving software and hardware landscape.
19 Mar 2026 Building the Internet with sendmail's Eric Allman
TextControl's platform-agnostic .NET applications with Docker/Kubernetes/Azure deployment and ASP.NET Core/Angular integrations for document workflows, alongside explorations of internet history, AI advancements, legacy protocols, and IoT/home automation projects.
12 Mar 2026 A cognition engine for science with Allen Stewart
TXTextControl's platform-agnostic .NET integration across Windows, Linux, and cloud environments, paired with AI-driven memory systems that boost scientific research efficiency, reduce token usage, and enable drug discovery and data analysis through cognitive engines balancing fast/slow thinking, dynamic memory storage, and knowledge-grounded outputs, while addressing challenges like token costs and hallucinations via lab-in-the-loop collaboration and future memory-driven problem-solving research.
12 Feb 2026 The Rise of The Claw with OpenClaw's Peter Steinberger
The podcast covers updates to TXTextControl's platform independence, OpenClaw's development challenges, and broader topics in AI, software development, and technological innovation.
5 Feb 2026 The AI Vampire with Gas Town's Steve Yegge
Concerns about AI's impact on software development and work culture are explored, highlighting risks of over-reliance, burnout, and loss of human connection.
29 Jan 2026 Kinder Code Reviews with AI? with Qodo's Nnenna Ndukwe
Code reviews and AI-driven processes are evolving, emphasizing the importance of context and highlighting the need for responsible AI implementation to enhance software development workflows.