From Tools to Agents: How Agentic Engineering is Reshaping Simulation-Driven Product Design

In collaboration with

Simulation is critical to engineering decisions, yet in many organizations it remains siloed, expert-led, and hard to scale across product development. As product complexity rises and development cycles compress, the challenge is no longer access to simulation. It’s how engineering work is coordinated and decisions are made with confidence.

In this session, Jon Wilde, VP of Product at SimScale, introduces agentic engineering workflows: AI-driven systems that autonomously or semi-autonomously guide engineering tasks, automate simulation setup and handoffs, and scale expert methods across teams. Built on SimScale’s cloud-native platform and grounded in physics-based simulation, these workflows bring together Engineering AI and Physics AI to determine when simulation is required, select appropriate models, and learn from results, transforming simulation from a specialist bottleneck into a scalable, system-level capability.

Armin Narimanzadeh, Manager of Thermofluids & Simulations at Convion, then shows what this looks like in practice. Convion, a subsidiary of HD Hydrogen developing solid oxide fuel cell and electrolyzer systems for the hydrogen economy, used SimScale’s Physics AI to compress a design optimization cycle from months to under one hour. The team evaluated thousands of design variants and identified a non-intuitive geometry that halved physical volume while meeting all performance targets. Armin walks through the workflow and explains how a validated AI model is now deployed as a shared tool across Convion’s engineering organization.

Watch this practical, interactive session to see how agentic engineering works and what it means for your workflows.

You will learn to:

• Implement agentic workflows to automate complex simulation tasks and scale expert-level engineering methods across your entire organization.
• Merge Engineering AI with physics-based simulation to accurately determine when to simulate and which models provide the most confidence.
• Compress design cycles from months to minutes by leveraging Physics AI to evaluate thousands of variants and identify optimized, non-intuitive geometries.