For Software Verification and Validation
Tasking Integrates AI Technology in its Toolchain
Tasking, provider of embedded software development tools, announced enhancements to the Tasking toolchain that enable seamless integration of AI in the software development and verification workflows.
The new capabilities accelerate the design and increase the performance of functionally safe and secure embedded real-time applications within automotive, aerospace & defense, industrial, and robotics while enabling OEMs to verify and validate (V&V) systems using agentic AI workflows.
Previously, workflows, processes, and tests were designed by hand. Developers can now use large language models (LLMs) to direct external AI agents and tools to automate many repetitive manual design, debug, and testing tasks. For example, AI-assisted workflows can implement comprehensive testing in earlier design stages to identify and resolve a wide range of issues sooner. The result is faster time-to-market with less chance of human error, greater overall system reliability, and lower development investment.
AI-assisted workflows can implement comprehensive testing in earlier design stages to identify and resolve a wide range of issues sooner, resulting in faster time to market with less chance of human error, greater overall system reliability, and lower development investment.
“AI enables today’s developers to be both more productive and efficient while at the same time delivering higher software performance and quality,” said Christoph Herzog, co-CEO of Tasking. “By taking over tedious and time-consuming tasks, AI-assisted tools can free up individuals to focus on value-added design. With the AI capabilities of the TASKING tools, development teams can now develop, verify, and validate complex systems faster, with less risk, and at a lower cost.”
Agentic AI Workflows
AI is playing an increasingly significant role in the design, debug, and test of real-time embedded applications. However, as AI is often implemented as a probabilistic tool, it can produce different results each time it is used. For systems that need to verify and validate deterministic behavior, AI must be carefully introduced to the workflow in a manner that enables systems to adhere to strict industry standards. For the foreseeable future, humans will remain in the development loop, so making them more productive and efficient in these processes is a difference maker to organizations willing to leverage AI.
The Tasking toolchain has been designed with a foundation that enables OEMs to develop functionally safe and secure systems. Modern AI capabilities are supported within the toolchain using the Model Context Protocol (MCP), an open-source standard that allows AI agents to securely interact within the development tools and access data required to achieve associated tasks. In this way, developers can use an LLM to control AI agents that direct and automate key aspects of the development lifecycle, helping teams create safer, more robust code. With this strategy, the developers can:
• Rigorously apply coding standards
• Optimize compilation configurations
• Automate Python-controlled debug tools
• Capture execution traces
• Manage iterative compile, debug, and test processes
• Verify that system specifications and requirements have been met
• Validate that code is performing the task for which it was written
• Simplify requirements traceability
The Tasking toolchain provides a rich array of reports that can be ingested by LLMs so they access accurate data about how code is structured and compiled, as well as how code executes on the target system (both virtual and physical). OEMs can utilize external AI resources set up in their own enterprises or in agentic development environments such as AWS Kiro, Microsoft Copilot, or Anthropic Claude Code.
“With the Tasking toolchain, AI can become an integral part of functionally safe and secure workflows,” said Janez Ulcakar, director of research & development, Tasking. “Workflows also become more flexible and agile, enabling developers to continuously optimize and enhance code with productivity enhanced by AI assistance. This gives OEMs the competitive edge of not just improving code design but of optimizing their entire software development lifecycle.”
Tasking’s overall compile – debug – test portfolio, including AI capabilities, will be showcased at embedded world 2026, in Hall 4, Booth 150. More information is available at https://www.tasking.com/events/ew26/.
