Clear zones for AI use
Teams leave with an agreed map of which repositories and tasks are suitable for agents. That avoids both blanket bans and uncontrolled use in safety-relevant code.
Vehicle and machine software is developed under process models, safety standards and supplier contracts that leave little room for experiments. Our workshop shows engineering teams where AI coding agents add value in this world and where human engineering judgment must stay in charge.
Project references from our team
+ 15,000 more
Automotive and industrial software teams rarely work on a single web app. They maintain embedded C and C++ under MISRA rules, AUTOSAR configurations, test automation for hardware-in-the-loop benches, MES integrations and a growing amount of cloud tooling for connected vehicles or machines. Requirements are traced from specification to test, and Automotive SPICE assessments check that the process is followed. An AI coding agent that writes code fast but cannot explain where a change came from creates more work than it saves.
That is why neooutput, the AI education unit of jocapps GmbH, separates the workshop into two zones. In tooling, test automation, data pipelines and internal applications, Claude Code, OpenAI Codex or Cursor can take on large parts of the work under review. In safety-relevant embedded code, the focus moves to explanation, test generation, static analysis findings and documentation, while qualified engineers keep the decisions. Sessions run on-site at your plant or development center or remote, in English or German.
Teams leave with an agreed map of which repositories and tasks are suitable for agents. That avoids both blanket bans and uncontrolled use in safety-relevant code.
Agents are well suited to writing test scripts, bench automation and test data generators. Test engineers practise this on examples close to their daily work.
We show how to link agent-assisted changes to requirement IDs, tickets and review records. Your ASPICE work products stay complete and consistent.
Agents can explain undocumented build scripts, code generators and old tool chains. That helps when experienced engineers retire or move to new platforms.
We discuss your product domain, safety level, languages, tool chain and current AI policy.
Together with your leads we classify repositories and tasks into suitable, restricted and excluded areas.
Engineers work with the agent on tooling, test automation and code explanation tasks using sample or approved code.
We document how agent use fits your review, traceability and qualification processes for a pilot.
Agentic coding workshops and AI services. Hands-on agentic coding workshops for teams on Claude Code, OpenAI Codex and Cursor, remote or on-site, in English or German. Plus AI Code Engineers on demand who deploy, secure and scale AI-built apps.
Explore neooutput workshopsRelated service: Agentic Coding Workshops
They can produce drafts, but in safety-relevant code we recommend using them mainly for explanation, review support and test generation. Final implementation decisions and verification stay with qualified engineers under your safety process.
We can include IEC 61131-3 code and automation interfaces if your teams work with them. Agent support for these languages varies, so we focus on documentation, test cases and code review rather than unsupervised generation.
Yes. On-site sessions are possible, which helps when code may not leave your network or when teams want to work directly with their own environments.
It depends on your IDEs, operating systems and security policy. We compare Claude Code, OpenAI Codex and Cursor against your setup during the scoping call.
15 minutes with an AI Code Engineer, free and without obligation. Or call our team around the clock.