AI Agent Designs Complete RISC-V Processor Core in Just 12 Hours

• April 23, 2026

Verkor.io, a startup focused on AI-driven chip development, has demonstrated a major advance in semiconductor design by employing an artificial intelligence agent to create a complete processor core based on the RISC-V architecture. According to a research paper released in March, the AI system, called Design Conductor, accomplished this task in a mere 12 hours, marking a significant acceleration compared to traditional chip design timelines.

Typically, commercial semiconductor development—from initial concept to a fully verified and layout-ready design—can take anywhere between 18 to 36 months. Verkor.io’s AI agent managed to condense this complex process into half a day, relying on only a short 219-word prompt as the initial specification.

AI-Driven Chip Design Breaks New Ground

Design Conductor’s output was a complete and verified microprocessor core compatible with the open-source RISC-V instruction set architecture. The AI took the brief prompt, which encapsulated design goals and key specifications, and autonomously generated a tested and ready-to-integrate schematic. This breakthrough highlights how artificial intelligence can efficiently handle intricate, multi-stage engineering workflows traditionally performed by human teams of engineers over many months.

The ability to rapidly design a processing unit not only accelerates innovation cycles but also has the potential to dramatically reduce costs and barriers to entry in the semiconductor industry. By automating much of the design validation and layout readiness, startups and established companies alike could shorten development schedules and respond to market needs more fluidly.

Verkor.io’s approach demonstrates the growing influence of AI beyond software automation into the realm of physical hardware design. While specific details regarding the performance characteristics of the AI-generated processor were not disclosed, the verified status of the design suggests a level of quality suitable for practical application.

This achievement also underscores the emerging trend of combining natural language processing prompts with complex engineering tasks. Using a concise written prompt of only a few hundred words, the AI was able to interpret design requirements accurately and convert them into functional hardware schematics, a process that traditionally relies heavily on specialized technical expertise and iterative refinement.

As semiconductor design continues to grow in complexity, AI-driven methodologies like Design Conductor could become essential tools in managing increasing demands for customization, power efficiency, and manufacturing viability. The industry’s shift towards harnessing AI agents for design-related challenges positions Verkor.io among the pioneers exploring the intersection of artificial intelligence and chip engineering.

Future research and development in this space will likely focus on expanding the capabilities of such AI agents, improving integration with existing electronic design automation tools, and validating the AI-generated designs in silicon. While the full impact remains to be seen, this development marks a promising step towards more agile and intelligent semiconductor design workflows.

Verkor.io’s AI system created a verified RISC-V processor core from a brief prompt in 12 hours, speeding chip design dramatically.