Project rules for your codebase
We write rules that capture architecture, naming, testing and forbidden patterns. Every developer and every agent run then follows the same conventions.
Most teams install Cursor in an afternoon and then use a fraction of it, each developer in a different way. Our training turns individual habits into a shared team setup with rules, context and review routines that make results predictable.
Project references from our team
+ 15,000 more
Cursor feels familiar because it builds on the VS Code experience, so adoption is fast. The difference between casual and effective use lies elsewhere: in project rules that tell the AI how your codebase works, in choosing between tab completion, inline edits and the agent for a given task, and in giving the agent the right files and documentation as context. Without shared conventions, one developer gets excellent results while another fights hallucinated APIs in the same repository.
neooutput, the AI education unit of jocapps GmbH, runs the training on your stack, remote or on-site, in English or German. We start with a short look at how your team uses Cursor today, then build rules for your main repositories, practise agent workflows on real backlog items and agree on how AI-generated changes are reviewed. Team leads also get guidance on privacy settings, model choice and how to keep usage costs transparent across the team.
We write rules that capture architecture, naming, testing and forbidden patterns. Every developer and every agent run then follows the same conventions.
Developers learn when tab completion, inline edits or the agent is the best fit. That saves time and avoids large unreviewed changes for small problems.
We practise referencing files, docs and symbols deliberately and keeping the index clean. The AI then works from your real code instead of guessing.
We review privacy mode, ignore files and model choices with your team lead. Sensitive files stay excluded and the setup is documented.
A short survey and call show how your developers use Cursor today and where results vary.
Together we write project rules and ignore files for your most important repositories.
Developers work on real tickets with the agent, compare approaches and review each other's results.
You receive a concise playbook with rules, review criteria and recommended settings for new team members.
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
Experienced users usually gain the most from shared project rules, better context selection and agent workflows for larger tasks. The training aligns individual tricks into one team standard.
Yes. If some developers prefer Claude Code or OpenAI Codex, we show how rules and conventions can be kept consistent across tools so the codebase does not drift.
We adapt to your stack, whether that is TypeScript, Python, Java, Go or something else. Exercises are based on your repositories or comparable samples.
Yes. We offer remote and on-site formats in English or German, and teams often combine an on-site day with remote follow-up sessions.
15 minutes with an AI Code Engineer, free and without obligation. Or call our team around the clock.