With Release 2026a, MathWorks is moving beyond AI-assisted coding to engineering-aware AI. In this exclusive interview, Prashant Saxena, Country Manager, MathWorks India, explains to Electronics For You’s Akanksha Sondhi Gaur how MATLAB and Simulink are evolving with copilots, agentic AI, and engineering-grounded workflows to accelerate product development while preserving the rigour demanded by safety-critical industries.
Q. Why is AI the focus of the 2026a release?
A. AI has become a foundational capability across engineering, helping engineers explore designs, develop algorithms, automate coding, and improve productivity. As MathWorks’ mission is to help engineers build better systems, MATLAB Copilot and Simulink Copilot, which are parts of Release 2026a, embed generative AI directly into MATLAB and Simulink. Additionally, MCP Server, MATLAB Agentic Toolkit, and Simulink Agentic Toolkit integrate Agentic AI into trusted engineering workflows. This ensures accuracy, repeatability, and trust while accelerating development from concept to deployment. Beyond AI, the release also introduces workflow enhancements that streamline research, design, testing, deployment, and operational monitoring, enabling faster innovation without compromising engineering rigour.
Q. What are the key innovations in Release 2026a, and how do they improve engineering workflows?
A. Release 2026a introduces AI-powered engineering assistants, MATLAB Copilot, Simulink Copilot, and Polyspace Copilot, to help engineers understand models, generate code, improve documentation, create test vectors, and accelerate development within familiar workflows. It also expands into agentic AI, enabling AI agents to generate code, run simulations, analyse results, and iteratively refine solutions until engineering goals are met. Complementing these are model context protocol (MCP) servers, engineering-specific AI toolkits, and workflow integrations that combine leading LLMs with MATLAB’s numerical computing and Simulink’s model-based design, advancing from AI-assisted to AI-enabled engineering.
Q. How do the AI capabilities differ from generic AI assistants?
A. Unlike general AI assistants, MathWorks’ AI capabilities are built around engineering knowledge rather than generic programming. MATLAB Copilot and Simulink Copilot understand mathematical models, physical systems, simulations, design constraints, and engineering best practices because they are integrated directly into MATLAB and Simulink. This enables context-aware assistance tailored to engineering tasks while allowing engineers to stay within their development environment, making AI support more accurate, efficient, and aligned with real engineering workflows.
Q. How do you balance AI-driven speed with engineering-grade accuracy and verification?
A. Generative and agentic AI can significantly boost engineering productivity, but AI-generated outputs must still be verified, validated, and traced to requirements, especially in safety-critical applications. MATLAB, Simulink, and Polyspace support these verification, validation, testing, and certification workflows, ensuring designs behave as intended. Rather than replacing engineering judgment, MathWorks combines AI-driven productivity with trusted engineering processes, enabling engineers to turn probabilistic AI outputs into deterministic, verifiable results while remaining in control.
Q. Beyond code generation, how does MATLAB Copilot improve engineers’ daily workflows?
A. MATLAB Copilot is designed as an engineering assistant, not just a coding tool. Beyond code generation, it explains existing code, generates documentation and comments, creates test vectors, and helps engineers understand unfamiliar algorithms, reducing time spent on repetitive tasks. Integrated directly into MATLAB, it provides context-aware assistance based on MATLAB’s capabilities and documentation, allowing engineers to work without leaving their environment. Its goal is to enhance, not replace, engineering expertise by freeing engineers to focus their time on system design, optimisation, and innovation rather than routine development activities.
Q. What does ‘grounded in engineering models and workflows’ mean in practice?
A. Simulink Copilot is built around the engineering knowledge and best practices embedded in the Simulink ecosystem rather than general programming. It understands mathematical models, physical systems, model-based design and workflows for embedded and control systems, enabling context-aware assistance tailored to engineering tasks. By aligning with established engineering methodologies, it delivers more relevant guidance, supports better design decisions, and integrates naturally into professional engineering workflows.
Q. How does Simulink Copilot simplify large, multi-team engineering projects?
A. One of the biggest challenges in large engineering projects is understanding complex legacy models developed by different teams, often with incomplete or outdated documentation. Before making changes, engineers must understand the architecture, trace signals and interpret past design decisions. Simulink Copilot shortens this learning curve by explaining model sections, component relationships and behaviour, tracing signals, identifying changes, and generating or improving documentation. By enabling conversational interaction with existing models, it helps engineers quickly gain context and become productive with greater confidence and efficiency.
Q. How will Agentic AI transform engineering workflows?
A. Traditional generative AI mainly generates text or code from prompts. Agentic AI goes further by executing commands, analysing results, refining its approach and repeating the process until the engineering goal is achieved. Instead of just answering questions, it can automate multiple development steps, such as writing code, running MATLAB commands, analysing simulations, and improving solutions, while keeping engineers in control to review and validate each stage. This combination of agentic AI and trusted engineering tools can significantly reduce repetitive work while maintaining engineering oversight.
Q. Why are MCP Servers and Agentic Toolkits important additions?
A. As engineers adopt LLMs such as Claude, Codex and Gemini, they need these AI systems to interact directly with engineering tools. MathWorks addresses this through its MCP Servers and open source MATLAB and Simulink Agentic Toolkits. The MCP servers provide a secure interface for AI assistants to communicate with MATLAB and Simulink, enabling them to execute simulations, analyse results, and complete engineering tasks within a controlled workflow. The GitHub-based toolkits add engineering-specific skills and workflows, giving developers a ready foundation for AI-assisted applications. Since they are updated through GitHub rather than product release cycles, new AI capabilities can be delivered much faster.
Q. Can you share a few practical examples of how engineers can use these new AI capabilities in their day-to-day work?
A. These capabilities simplify practical engineering workflows. For example, an engineer can describe a data analysis task in natural language, and the AI generates MATLAB code, executes it, creates plots, analyses the results and iteratively refines the solution until the objective is met. In Simulink, engineers can describe a system, such as a solar tracking controller, and the AI can generate the initial model, configure blocks, run simulations and help refine the design. The same approach extends to test generation, model refinement and documentation, enabling engineers to focus on optimisation and system-level decisions instead of repetitive implementation tasks.
Q. How significant is AI-generated system architecture for product development?
A. AI has the potential to transform the early stages of system design by helping engineers move directly from requirements to an initial system architecture. Instead of manually interpreting requirements documents, engineers can provide requirements as documents or images and describe the desired system in natural language. AI then generates an initial architecture in System Composer, which engineers review, validate and refine using their expertise. While it does not replace engineering judgement, this capability reduces manual effort, accelerates concept development, enables earlier evaluation of design alternatives and shortens the path from requirements to an executable system model, helping reduce overall development time.
Q. How are you supporting AI-assisted development in environments like Visual Studio Code?
A. Modern engineering workflows span multiple development environments, so our AI capabilities are designed to integrate seamlessly rather than create separate workflows. Through Agentic Toolkits and MCP Servers, AI assistants in environments such as Visual Studio Code can securely invoke MATLAB, run analyses, perform simulations, and return engineering results. This allows engineers to work in their preferred environment while leveraging MATLAB’s numerical computing and Simulink’s model-based design capabilities, creating a more flexible and seamless AI-assisted engineering workflow.
Q. How does Polyspace Copilot improve software quality?
A. As embedded software becomes more complex, early defect detection is critical to meeting quality and safety requirements. Polyspace Copilot combines AI-assisted interaction with Polyspace’s static code analysis, allowing engineers to conversationally explore verification results, understand issues and identify appropriate fixes. Complementing this, Polyspace As You Code brings verification directly into the development environment, enabling engineers to detect and correct defects while writing code instead of later in the lifecycle. Together, these capabilities improve software quality while helping teams maintain productivity.
Q. How do you ensure AI-generated code meets engineering quality standards?
A. AI can accelerate software and model development, but AI-generated outputs must still undergo rigorous verification, validation and testing, especially in safety-critical industries such as automotive, aerospace, medical devices, and industrial automation. MATLAB, Simulink, and Polyspace provide established workflows for simulation, testing, static analysis, traceability, and certification support, with AI complementing, not replacing, these capabilities. Our approach is to automate repetitive engineering tasks while relying on trusted verification workflows to build confidence in the final design. Engineers remain responsible for validating results and ensuring compliance with industry standards.
Q: Which industries will benefit most from Release 2026a’s AI capabilities?
A. AI has applications across almost every engineering-intensive industry by reducing repetitive work, accelerating design iterations and improving productivity, although adoption will vary with regulatory and safety requirements. Automotive is expected to lead due to software-defined vehicles, ADAS, electrification and connectivity, while semiconductor, aerospace, and defence companies can use AI to accelerate development without compromising verification or certification. Industrial automation, robotics, communications, energy, and medical devices will also benefit as AI helps engineers manage growing system complexity, automate routine tasks, and evaluate more design alternatives. Across all industries, AI strengthens engineering processes rather than replacing them, enabling engineers to focus on innovation while trusted workflows ensure quality and compliance.
Q. What business benefits can engineering leaders expect from Release 2026a?
A. Engineering leaders face increasing pressure to deliver more sophisticated products in shorter development cycles. AI adds value by improving the efficiency of the entire product development process, not just automating individual tasks. Integrated into MATLAB and Simulink, it reduces time spent on routine coding, understanding legacy projects, generating documentation, and creating initial models, allowing engineers to focus on optimisation, innovation and validation. AI also enables teams to explore more design alternatives within the same schedule by quickly generating and refining initial solutions. At the same time, engineering-specific AI workflows help optimise computational costs by reducing unnecessary token usage, delivering higher productivity, better engineering quality and faster development.
Q. How important is India to your long-term strategy?
A. India has evolved into one of the world’s leading engineering innovation hubs, with major automotive, semiconductor, aerospace, industrial and technology companies using their Indian engineering and Global Capability Centre (GCC) teams for core product development. Since establishing a direct presence in India in 2008, MathWorks has seen these teams progress from supporting global operations to making architectural decisions, developing complex algorithms and building software-defined products. As MATLAB and Simulink support the entire product development lifecycle, this growing engineering responsibility is driving increased demand for integrated workflows and AI-enabled productivity, making India a key strategic global market.
Q. How are Indian engineering organisations adopting AI-driven workflows?
A. We are seeing strong interest from Indian engineering organisations, which recognise AI’s potential to improve productivity and are evaluating it through proof-of-concept projects and pilot deployments. Adoption varies by application, with companies developing safety-critical products taking a cautious approach because AI-generated outputs still require rigorous verification and validation. Even so, they are actively exploring AI for documentation, model understanding, code assistance, and workflow automation, where productivity gains can be achieved without compromising engineering integrity. Combined with India’s growing engineering talent and increasing ownership of global product development, this positions the country for wider adoption of AI-assisted engineering in the years ahead.
Q. How are you preparing future engineers for AI-driven engineering?
A. Preparing future engineers requires teaching not only how to use AI tools but also how to apply them within sound engineering practices. MathWorks works closely with universities to give students hands-on experience with the same tools and workflows used in industry, while integrating AI into engineering education. The focus is on helping students understand how AI improves productivity alongside the continuing importance of verification, validation, and engineering judgement. This equips graduates with the skills needed to contribute effectively to AI-assisted engineering teams and drive future innovation.
Q. How will AI reshape the role of engineers?
A. We see AI as an enabler, not a replacement for engineers. Engineering requires solving complex real-world problems, making design trade-offs and ensuring systems operate safely and reliably, responsibilities that remain with engineers. AI can automate repetitive tasks such as writing routine code, generating documentation, analysing models, creating test cases, and setting up simulations, freeing engineers to focus on innovation, system architecture, and critical design decisions. As AI advances, engineering expertise will become even more important, with engineers guiding AI, validating its outputs and ensuring compliance with performance, safety, and regulatory requirements.
Q. What skills will engineers need to thrive in an AI-driven future?
A. Strong engineering fundamentals will remain the most important skill, as AI cannot replace a deep understanding of mathematics, physics, control systems, embedded software, and domain-specific engineering principles. This knowledge enables engineers to evaluate whether AI-generated solutions are technically sound. At the same time, engineers should develop AI literacy by learning how to frame engineering problems, interpret AI-generated results, and integrate AI into existing workflows. As AI adoption grows, expertise in verification, validation, and traceability will become even more important to ensure AI-generated outputs meet engineering standards. The engineers best positioned for the future will combine strong domain knowledge with the ability to work effectively alongside AI.
Q. How will AI transform engineering in the coming years?
A. We believe AI will become a natural part of the engineering development environment rather than an external productivity tool. Engineers will increasingly work with intelligent assistants and autonomous agents that understand engineering objectives, run simulations, analyse results, and recommend improvements throughout the product development lifecycle. However, AI will complement, notreplace, trusted engineering workflows built on simulation, testing, verification, and validation. By enabling a workflow that combines AI with MATLAB and Simulink, our goal is to help engineers innovate faster while maintaining the highest standards of quality, reliability and safety.
Q. What message do you have for engineering leaders evaluating AI?
A. AI should be viewed as a strategic engineering capability, not just another software feature. When integrated thoughtfully into engineering workflows, it can improve productivity, shorten development cycles, and free engineers to focus on higher-value problems. At the same time, successful adoption depends on maintaining rigorous verification, validation, traceability, and engineering judgement, especially for safety-critical applications. Our vision is to combine the speed and flexibility of AI with trusted engineering workflows, making AI a catalyst for better engineering rather than simply a coding assistant, and Release 2026a is an important step towards that future.
