Developers can now use the LLM and agentic AI of their choice to automate software development in a way that supports requirements traceability and simplifies certification to functional safety standards

TASKING, a global leader in high-performance embedded software development tools, has launched the winIDEA AI Kit. This set of tools and files allows artificial intelligent (AI) agents to integrate with TASKING’s unified toolchain through the winIDEA development environment. Accessibility to AI agents within the development toolchain enables developers to automate multiple development workflows, accelerating software design and verification while maintaining strict requirements traceability to facilitate and simplify certification.

Embedded teams, especially those developing complex, multicore applications across automotive, aerospace and defense, industrial, medical, and robotics, face the challenge of delivering increasingly complex, functionally safe software under tighter development timelines. While AI can be used to help write, test, and verify software, requirements traceability requires developers take full responsibility for all AI activity. TASKING integrates AI into software workflows and provides context to generically trained LLMs on how code/scripts should be generated to improve developer productivity while maintaining the requirements traceability needed for certification to industry standards. Guardrails define what AI can do, while human oversight ensures accountability by enabling AI to support the development lifecycle without compromising reliability, functional safety or security.

“Our approach to AI accelerates complex, multicore embedded development by giving agents structured access to debug and analyze workflows,” said Janez Ulcakar, Director of R&D Debugger Solutions, TASKING. “This enables agentic-based automation to work with real-time debug evidence while ensuring engineers remain in full control of AI processes and conclusions. This is essential for systems that need to be certified to functional safety standards.”

The winIDEA AI Kit at a Glance

TASKING’s winIDEA AI Kit combines three maintained components designed to simplify and accelerate embedded development workflows:

• winIDEA MCP Server: Provides structured access to design tool capabilities, including debug control, memory, variables, breakpoints, trace, coverage, profiling, and timing analysis
• AI-ready winIDEA SDK Documentation: Assists agents in generating accurate automation scripts in supported programming languages
• Embedded-Development Skills: Offers task-oriented guidance and context for debugging, timing analysis, winIDEA configuration, and device-specific workflows that draw upon TASKING’s extensive embedded software development experience and best practices

With winIDEA AI Kit, teams can ask an AI agent to perform tasks instead of creating and maintaining an extensive library of custom scripts or tool instructions. The kit also aims to improve the accuracy of SDK-generated scripts. During a live debug session, AI agents can assist with tasks such as:

• Pinpointing the exact location of an application crash
• Analyzing CPU load and identifying which OS tasks consume the most CPU resources
• Inspecting variables, memory, processor states, and breakpoints
• Configuring and analyzing trace, timing, and code coverage
• Executing supported testing and analysis workflows and working with their results

AI agents are able to help automate even the most complex tasks across the entire tool chain. For example, AI agents can significantly simplify timing coupling analysis in multicore systems by automating execution of different task combinations across cores to evaluate worst-case execution timing (WCET). This process involves automating the compiler, debugger, and test tools to work together without requiring manual intervention. The result is greater timing coupling coverage and more accurate assessment of WCET faster and with reduced risk of human error.

“At TASKING, we see AI as a tool to make engineers more productive, not replace them,” said Christoph Herzog, Co-CEO and CTO, TASKING. “When AI is founded on helping automate complex processes, developers can step away from tedious, repetitive tasks and focus more on increasing value for the customer.”

As the winIDEA AI Kit is model- and IDE-agnostic, the AI experience is not tied to one assistant or debugger. This gives developers the flexibility to build agentic workflows using the LLM and agentic IDE of their choice with maintained, domain-specific tools. TASKING maintains the AI Kit alongside winIDEA releases so teams can adopt AI to fit their particular tools, workflows, and data policies. The kit supports cloud-based models and locally deployed LLMs for environments without internet access.

Developers can learn more about how TASKING’s winIDEA AI Kit is changing the way the world develops embedded software at https://www.tasking.com/campaign/winidea-ai-kit/. For more information on use cases, see Agentic AI Debugging with TASKING winIDEA at https://www.youtube.com/watch?v=R4J-KP74QSQ.

About TASKING

TASKING is a leading provider of embedded software development tools that compile, debug and test. These tools enable embedded software engineers to develop reliable, high-performance applications for a safer future. Founded in 1977, TASKING is committed to expanding its portfolio to deliver a single-supplier, certifiable tool chain. As a trusted partner focused on customer need, expertise and a commitment to sustainability, TASKING and its integrated tool chain accelerate the software development lifecycle for safety and security critical applications in industries such as automotive, aerospace and defense, industrial, medical, robotics and others.

For more information, visit our website or follow us on LinkedIn.

View source version on businesswire.com: https://www.businesswire.com/news/home/20261014110111/en/

TASKING’s new winIDEA AI Kit accelerates software design and verification while maintaining strict requirements traceability to facilitate and simplify certification.