AMD has 'supercharged' its software development by using AI agents. These agents run in automated loops, constantly analyzing and optimizing customer models for AMD's hardware. This turns a slow, manual process into a scalable, nonstop operation, dramatically improving out-of-the-box performance for developers.

Previously, building bespoke software for niche internal problems was too expensive. AI agents dramatically lower this cost, allowing companies to create custom-fit solutions for 99% of their problems, ending the era of contorting workflows to fit generic, off-the-shelf tools.

Advanced AI like Astra dramatically lowers the barrier to creating highly custom software. Projects that were previously too complex for individuals—like reverse-engineering proprietary hardware or building a retro UI wrapper—can now be generated quickly, enabling a new wave of personalized applications.

Hardware vendors like NVIDIA (CUDA) and AMD create fragmented, proprietary software stacks that lock developers in. Modular builds a replacement layer that enables AI models to run consistently across different hardware, giving enterprises choice and flexibility without rewriting code.

The technical friction of setting up AI agents creates a market for dedicated hardware solutions that abstract away complexity, much like Sonos did for home audio, making powerful AI accessible to non-technical users.

Nvidia's CUDA software has created a powerful developer lock-in. However, the advancement of AI coding agents is weakening this moat. These agents can automate the difficult process of writing performant code for competing, non-CUDA chipsets, reducing the switching costs for AI labs.

The era of dual-purpose AI chips is ending. The overwhelming demand for real-time processing from AI agents is forcing companies like Google and NVIDIA to create dedicated, inference-optimized hardware. This marks a fundamental and permanent split in the AI infrastructure market, separating training from inference.

The rise of agent orchestration using specialized, open-source models will drive demand for custom ASICs. Jerry Murdock argues that putting a model on a dedicated chip will be far cheaper and more tunable for specific workloads than using expensive, general-purpose GPUs like Nvidia's, spurring a hardware shift.

Wafer's core innovation is using specialized AI agents to perform the complex, menial task of optimizing LLMs to run efficiently on GPUs. This automates the work of the ~500 highly-paid, scarce GPU kernel engineers in the world, dramatically accelerating performance optimization.