A profound transformation is reshaping the engineering world. With the emergence of AI agents, machines are no longer confined to the role of passive tools: they are becoming active partners in reasoning, analysis and decision-making, explains Cédric Kalifa, VP Sales WER & France Country Manager, PTC.
Especially in highly complex and regulated industries such as mechanical engineering, automotive and aerospace industries, this shift represents a true paradigm change. Isolated workflows are gradually giving way to interconnected, modular systems that automate repetitive tasks while providing meaningful support to development teams.
From Automation to True Assistance
AI in engineering is not a new topic. What is new, however, is the rise of agentic AI, autonomous, contextual and goal-oriented, which is fundamentally redefining expectations. The challenge is no longer to automate isolated work steps, but to understand and support entire processes from end to end.
AI agents learn from data, detect correlations, generate recommendations and act according to predefined objectives. They do not replace engineers. Instead, they assist them where traditional approaches reach their limits, whether in analyzing complex system architectures, evaluating variants, ensuring regulatory compliance or managing the growing number of configurations.
This marks a shift away from automation in its traditional sense toward a new form of digital assistance: integrated, traceable and expendable.
Why New Solutions Have Become Essential
The pressure on engineering teams is steadily increasing. Product variants are multiplying, innovation cycles are shortening, regulatory constraints are intensifying and expectations around quality and customization continue to rise. Together, these factors are pushing traditional engineering processes to their limits.
The impact is particularly visible in the context of software-defined vehicles (SDVs), where development now relies on data, interfaces and the ability to update systems in real time.
This is where the cracks begin to show. Many development workflows still depend on document-based approvals, manual review processes and poorly connected tools. The result is a lack of continuity, version conflicts, extended feedback loops and ultimately, slower innovation exposed to a higher risk of errors.
A Structured Response to Engineering Complexity
AI agents offer a structured response to these challenges. They are taking on new roles within modern PLM, ALM and CAD environments. By analyzing requirements, test reports and configuration states, they identify patterns that reveal risks, opportunities or the need for adjustments. From there, they generate recommendations for action, prioritize tasks and highlight potential conflicts.
In model-based engineering environments such as MBSE (Model-Based Systems Engineering), AI systems can evaluate multiple design options and simulate their impact on performance, compliance and cost even before any physical prototype is built. This shortens iteration cycles and allows risks to be identified at an early stage in the development process.
In fast-moving projects, AI agents automatically evaluate how planned changes will affect existing requirements, components, or tests. They suggest implementation paths and assess regulatory risks, which supports informed decision-making.
For safety-critical systems such as vehicles or aircraft, full traceability is non-negotiable. AI agents verify the feasibility of new configurations, their compliance with standards and the completeness of the documentation required to demonstrate that compliance. The payoff is significant: time saved, resources optimized.
The Intelligent Product Lifecycle as a Foundation
To deliver value, AI agents require a strong data backbone. This is where the Intelligent Product Lifecycle (IPL) comes into play: a structured, semantically consistent data architecture that links all relevant information across the entire lifecycle of a product.
From initial requirements to end-of-life, from software versions to maintenance history, the IPL enables AI to operate not only locally, but systemically.
It ensures continuous traceability, automated version management and cross-functional transparency between teams and tools. It also enables context-sensitive assistance through AI agents and supports faster, well-documented decisions aligned with regulatory processes – a decisive advantage in safety-critical industries.
Practical Examples From Industry
Agentic AI is already delivering tangible results across industrial sectors. Several examples demonstrate its impact:
- Control unit development: In the automotive industry, AI agents analyze consistency between software versions, requirements and test cases. At BMW, for example, AI is used to automatically check variant combinations for plausibility and approval eligibility.
- Aerospace cable manufacturing: AI systems detect irregularities in color, insulation or diameter and compare this data in real time with the product’s digital twin. This allows errors to be detected and corrected at an early stage.
- Mechanical engineering: In plastic injection molding, AI-based validation tools reduce the number of iterations required to reach series production by up to 40%, significantly improving quality, lead times and resource efficiency.
Engineering Reimagined: Humans as Orchestrators
These advances are reshaping roles within engineering teams. Design engineers, system architects and test managers are increasingly becoming orchestrators of digital processes. They define rules, maintain semantic models, configure and train agents, and interpret their recommendations in the overall context.
This new role requires more than traditional engineering expertise. It calls for systemic thinking and domain knowledge, confident handling of data and tools, and the ability to collaborate between humans and AI systems. Only then can the full potential of agent-based support be realized.
Building Trust Through Transparency and Traceability
Transparency is a key factor in the adoption of agentic AI in engineering. Decisions made by AI agents must be traceable and verifiable for developers, but also for audit and management teams. A transparent IPL, clear rules and versioned decision trees are essential here.
To scale in environments where safety, quality and compliance are critical, agentic AI cannot function as a “black box.” It must operate as an explainable, well-documented decision-support system.
Agentic engineering is more than a technical trend. It represents a structural transformation. Organizations that integrate AI agents into their toolchains today gain operational efficiency and, more importantly, lay the foundations for scalable, resilient and future-proof development processes.
The next generation of digital product development will be built on close collaboration between humans and AI. It combines the best of both worlds: the precision of machines and the creativity and responsibility of human engineers. And that, in the end, is the real revolution: engineering is once again becoming a place of genuine innovation.
