Uploaded August 2026 | Updated September 2026, 2 weeks ago
Artificial intelligence is influencing how engineers design, develop, verify, and validate complex systems. In this presentation, Mehran Mestchian, who leads control design automation and verification technology development at MathWorks, examines how AI is being integrated into engineering workflows through embedded intelligence, generative AI, and agent-based design.
The talk outlines the progression from traditional machine learning methods toward AI systems that support iterative, goal-directed design processes. These systems can generate candidate solutions, evaluate results, and refine designs in closed-loop workflows. Rather than replacing established engineering practices, these approaches extend principles such as data-driven development, simulation, testing, and optimization.
A key theme is the integration of AI into engineering design loops while keeping production code generation deterministic. The presentation illustrates how generative and agent-based AI can support earlier design activities—such as concept exploration, model creation from requirements, tradeoff studies, and evaluation-loop setup—before validated models are used to generate traceable, qualifiable production code. It emphasizes that trust depends on rigorous evaluation through simulation, validation, benchmarking, testing, and quality checks.
The session explores how Model-Based Design evolves as generative AI and agent-based AI become part of engineering work. The emphasis is on where AI is useful: helping engineers explore options, structure design tasks, and run evaluation cycles at a higher level of abstraction. Engineers still define the goals, constraints, guardrails, and acceptance criteria, while simulation, testing, validation, and verification provide the evidence needed to decide whether an AI-suggested result is usable.
Watch to gain insight into how workflows are evolving as AI enters engineering work: where agent-based design can accelerate iteration, where deterministic and verifiable methods remain essential, and where AI can be adopted without weakening engineering discipline.
Learn more about Generative AI with MATLAB and Simulink: bit.ly/4foLM8P
Chapters:
00:00 How AI Is Changing Engineering Design
02:17 MATLAB, Simulink, and Neural Networks
04:50 Four AI Epochs from Research to Generative AI
06:27 How AI Reimagines Engineering Design Loops
09:36 Agentic AI and Engineering Workflow Automation
12:53 Probabilistic AI vs. Deterministic Engineering Systems
14:07 Model-Based Design and Generative AI
20:46 AI with Model-Based Design, MCP Servers, and Simulink
--------------------------------------------------------------------------------------------------------
Get a free product trial: goo.gl/ZHFb5u
Learn more about MATLAB: goo.gl/8QV7ZZ
Learn more about Simulink: goo.gl/nqnbLe
See what's new in MATLAB and Simulink: goo.gl/pgGtod
© 2026 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc.
See mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be trademarks or registered trademarks of their respective holders.
Artificial intelligence is influencing how engineers design, develop, verify, and validate complex systems. In this presentation, Mehran Mestchian, who leads control design automation and verification technology development at MathWorks, examines how AI is being integrated into engineering workflows through embedded intelligence, generative AI, and agent-based design.
The talk outlines the progression from traditional machine learning methods toward AI systems that support iterative, goal-directed design processes. These systems can generate candidate solutions, evaluate results, and refine designs in closed-loop workflows. Rather than replacing established engineering practices, these approaches extend principles such as data-driven development, simulation, testing, and optimization.
A key theme is the integration of AI into engineering design loops while keeping production code generation deterministic. The presentation illustrates how generative and agent-based AI can support earlier design activities—such as concept exploration, model creation from requirements, tradeoff studies, and evaluation-loop setup—before validated models are used to generate traceable, qualifiable production code. It emphasizes that trust depends on rigorous evaluation through simulation, validation, benchmarking, testing, and quality checks.
The session explores how Model-Based Design evolves as generative AI and agent-based AI become part of engineering work. The emphasis is on where AI is useful: helping engineers explore options, structure design tasks, and run evaluation cycles at a higher level of abstraction. Engineers still define the goals, constraints, guardrails, and acceptance criteria, while simulation, testing, validation, and verification provide the evidence needed to decide whether an AI-suggested result is usable.
Watch to gain insight into how workflows are evolving as AI enters engineering work: where agent-based design can accelerate iteration, where deterministic and verifiable methods remain essential, and where AI can be adopted without weakening engineering discipline.
Learn more about Generative AI with MATLAB and Simulink: bit.ly/4foLM8P
Chapters:
00:00 How AI Is Changing Engineering Design
02:17 MATLAB, Simulink, and Neural Networks
04:50 Four AI Epochs from Research to Generative AI
06:27 How AI Reimagines Engineering Design Loops
09:36 Agentic AI and Engineering Workflow Automation
12:53 Probabilistic AI vs. Deterministic Engineering Systems
14:07 Model-Based Design and Generative AI
20:46 AI with Model-Based Design, MCP Servers, and Simulink
--------------------------------------------------------------------------------------------------------
Get a free product trial: goo.gl/ZHFb5u
Learn more about MATLAB: goo.gl/8QV7ZZ
Learn more about Simulink: goo.gl/nqnbLe
See what's new in MATLAB and Simulink: goo.gl/pgGtod
© 2026 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc.
See mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be trademarks or registered trademarks of their respective holders.
