AI Engineer - Working Student

Role overview

Responsibilities

• check_circle Develop AI-first applications and MCP integrations that give our agents access to the tools and data they need to be genuinely useful
• check_circle Build RAG pipelines, internal agents, and venture-wide AI tooling that reaches across our portfolio
• check_circle Integrate AI platforms and services into our existing workflows so they get used day-to-day
• check_circle Implement AI-enabled features such as guided search and intelligent discovery
• check_circle Prototype new AI features quickly and turn the promising ones into production-ready code
• check_circle Measure and improve LLM output quality: set up evals, observability, and dashboards so we know when things drift or regress
• check_circle Help mature our LLMOps practice: prompt versioning, CI/CD, Docker, monitoring
• check_circle Explore new models, APIs, and patterns, and bring back what works
• check_circle LLM providers: Claude, OpenAI, Gemini, or through routers like OpenRouter
• check_circle Agent & orchestration frameworks: Mastra, LangChain, LlamaIndex, Semantic Kernel, or similar
• check_circle Evals & observability: Langfuse, DeepEval, or similar
• check_circle Web research & data extraction: Firecrawl, Tavily, Exa, or similar
• check_circle Cloud platforms: Azure, AWS, or GCP
• check_circle Data & databases: Databricks, Snowflake, PostgreSQL
• check_circle Vector search: Weaviate, Pinecone, Qdrant, Chroma
• check_circle Languages: Python (FastAPI, Flask, or Django) and/or TypeScript (Next.js, Express, or NestJS)
• check_circle You're a student in the final two years of an AI, Computer Science, or similar study
• check_circle You can show your passion: point us to work experience, projects, or anything else that proves you build AI things because you want to, not just because it's assigned
• check_circle You are curious and love to explore. You read release notes for fun, try out new models the day they drop, and would rather spend an afternoon experimenting with a new tool than waiting for someone to tell you what to use.
• check_circle You have an ownership mindset. You take a problem from "fuzzy idea" to "working thing" without needing someone to map every step. When something breaks, you dig in instead of waiting.
• check_circle You are a strong communicator. You can explain what you built and why: to both engineers and non-technical stakeholders, in writing and in conversation. You ask good questions and don't pretend to understand things you don't.
• check_circle You are a rapid prototyper. You ship a rough working version on day one and improve it from there. You know when "good enough" is good enough, and when something needs more care.
• check_circle Have built and shipped an end-to-end LLM project: a RAG system, agent, or pipeline you can walk us through and defend the design choices on
• check_circle Hands-on with an agent framework (Mastra, LangChain, LlamaIndex, or similar) and have built or integrated an MCP server
• check_circle Experience with evals tooling (Langfuse, LangSmith, or a homegrown setup) and comfortable writing test cases for non-deterministic outputs
• check_circle Some exposure to cloud platforms (Azure preferred, but AWS/GCP is fine)
• check_circle Interest in data engineering, RAG, vector search (Weaviate or similar), agents, or evaluations
• check_circle You've shipped at least one AI feature or internal tool that's actually being used
• check_circle You've set up evals or an observability dashboard for one of our LLM-powered workflows
• check_circle You're comfortable navigating our stack end-to-end
• check_circle You've contributed to at least one venture-wide solution
• check_circle The team trusts you to own a small piece of the product from idea to production

Benefits

• check_circle A fair and good compensation package of 3200 euros gross per month on a full-time basis
• check_circle 8% holiday allowance
• check_circle 28 annual holidays on a full-time basis
• check_circle Commuting allowance and work-from-home allowance
• check_circle Team lunches every Monday
• check_circle Personal development plan with your growth and development as important pillar
• check_circle A company standard guest house
• check_circle A part-time role that genuinely fits around your studies or other commitments
• check_circle Real ownership of features you'll see in production
• check_circle A small team to learn from and grow with
• check_circle Intro call: a casual 30-minute conversation to get to know each other and talk through what you've built
• check_circle Take-home assignment: a small, focused AI engineering task you can do in a few hours; we'll discuss your solution together
• check_circle Decision: We get back to you with a decision within 5 working days
• check_circle Flexibele werkuren
• check_circle Kosteloos parkeren
• check_circle Lunchkorting
• check_circle Pensioen
• check_circle Reiskostenvergoeding
• check_circle Telefoonplan
• check_circle Werk vanuit huis