TASKING Integrates Modern AI Technology to Enable Robust Software Verification and Validation (V&V)  

The TASKING tool­chain now sup­ports agen­tic AI work­flows that allow OEMs and Tier 1 sup­pli­ers to fur­ther opti­mize the entire soft­ware develop­ment life­cy­cle of func­tion­al­ly safe and secure sys­tems 

Munich – Feb­ru­ary 24, 2026 – TASKING, a glob­al leader in high-per­for­mance embed­ded soft­ware develop­ment tools, today announced enhance­ments to the TASKING tool­chain that enable seam­less inte­gra­tion of AI in the soft­ware develop­ment and ver­i­fi­ca­tion work­flows. These 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 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.” 

Learn how the TASKING tool­chain is chang­ing the way the world devel­ops embed­ded soft­ware. The over­all com­pile – debug – test port­fo­lio, includ­ing AI capa­bil­i­ties, will be show­cased by TASKING at Embed­ded World 2026, in Hall 4, Booth 4-150. For more infor­ma­tion, visit here.

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 tool­chain. As a trust­ed part­ner focused on cus­tomer need, exper­tise and a com­mit­ment to sus­tain­abil­i­ty, TASKING and its inte­grat­ed tool­chain accel­er­ates 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.

Media con­tacts:

Eylul Kocak, Mar­ket­ing Man­ag­er, TASKING
Tel:  +49 89 262010082, Email: eylul.kocak@tasking.com

Kelly Wan­lass, Media Rela­tions, HCI Mar­ket­ing and Com­mu­ni­ca­tions, Inc.
Tel: +1 (801) 602-4723, Email: kelly@hcimarketing.com

Scroll to Top