What this is about
You test and optimise our existing systems: platform features, internal tools, integrations, automations. The core idea: you work with Claude Code and similar AI coding tools. They make a Werkstudent 3–5x more productive, if you know how to drive them. We will show you how.
We are CegTec, GTM Engineering partner for premium B2B brands in the DACH region. Our outbound platform integrates HubSpot, Instantly and our own sales system. You work on a productive SaaS codebase, not a toy project.
20h/week during the semester, up to 40h during semester breaks. After graduation there is a clear path to a full-time engineering role.
What you do
Testing and bug triage (~40%)
- Write end-to-end tests for platform features, with Claude Code as a pair programmer
- Reproduce bug reports from operators, analyse root cause
- Regression tests for critical pipelines (reply triage, sending, CRM sync)
System optimisation (~30%)
- Identify and fix performance bottlenecks (DB queries, API latency, memory usage)
- Code refactoring based on senior engineer briefings
- Read logs, find anomalies, introduce cleaner logging patterns
Internal tools and automations (~20%)
- Build small internal tools that save operators or founders time (Claude Code as scaffolding)
- Integrations between tools (CRM ↔ cegtec, Slack ↔ reporting, n8n flows)
Documentation (~10%)
- Maintain code documentation
- Update runbooks for operations processes
What you bring
Must have:
- Enrolled student (CS, business informatics, or with demonstrable programming experience)
- Solid programming basics: one backend language (ideally TypeScript or Python), Git, command line, SQL basics
- Experience with AI coding tools (Claude Code, Cursor, GitHub Copilot), you know what you can delegate and what you cannot
- Self-direction, you ask when unclear but do not freeze
- Bug detective mindset, you chase a problem down to its root
Nice to have:
- First open-source contributions or your own projects with real users
- Experience with our stack (TypeScript/Node, Astro, n8n, Supabase, Postgres)
- Interest in SaaS systems and scaling
How we measure success
| Timeline | Expectation |
|---|---|
| After 30 days | Onboarding done, local dev environment running, first small PRs merged |
| After 90 days | Independent bug reproduction and fixes, first internal tool built and in use |
| After 180 days | Independent feature implementation from briefing, code review contributions to colleagues |
What you get
- €20/h, slightly above market for engineering Werkstudenten
- 20h/week during the semester, up to 40h during semester breaks
- Remote-first with an optional office in Hamburg
- Equipment and tool stack: laptop, all licences (Claude Code Max, GitHub, Cursor, all cloud accounts)
- What you learn: a productive SaaS codebase from the inside, and how to drive AI coding tools at senior level. This is the skill for the next ten years.
- Clear path to a full-time engineering role after graduation
How to apply
Send us with your application:
- Short CV
- Three sentences on why exactly this role
- Link to a code repository you built yourself (or a contribution to open source)
- Practical snippet: An operator reports that a reply triage pipeline has produced no new entries since last night. Logs show
ERR_RATE_LIMITon an external API call. In 5 sentences, describe how you would identify the root cause and apply a short-term fix.