Studocu helps millions of students find and share the study materials they need. We have 50M+ monthly active users, and we're building Studocu AI: a set of AI-powered features that make studying faster and more effective. Think course companions, diagnostic exams, and smart content recommendations, all powered by LLMs and agentic systems.

The Role

We're looking for a working student who wants to get their hands dirty building AI features that real students will use. You won't be fetching coffee or organizing spreadsheets. You'll be writing Python, working with LLMs/Langfuse/Langgraph and shipping code alongside our AI Engineering and Data Science team.

This is a role for someone who's already tinkering with AI on their own: building side projects, experimenting with APIs, breaking things and figuring out why. We care less about your grades and more about what you've built. This opening is strictly not an internship.

What you'll actually do

You'll work on real features within Studocu AI (studocu.com/ai). Depending on your skills and what's in flight, that could look like:

• Building and testing prompt chains and agent workflows in Langgraph

• Prototyping new AI features (evaluation tools, content processing pipelines, conversational flows)

• Running evaluations on model outputs writing scoring logic, comparing approaches, spotting failure modes

• Helping improve our document processing and retrieval pipelines

• Contributing to internal tools that help the team move faster

You'll be paired with a Senior AI engineer who will review your code, give you context, and make sure you're learning, but you'll be expected to own your work and push it forward.

Must-haves:

• Eligible to work in the Netherlands. We do not provide any sponsorships for this role.

• Currently enrolled in a Bachelor's or Master's programme (Computer Science, AI, Data Science, Math or similar)

• Comfortable writing code 

• Familiar with LLM APIs (OpenAI, Anthropic, or similar). You've called one from code, not just used ChatGPT

• You've built something with AI outside of coursework. A side project, a hackathon entry, a tool you made for yourself

• Available for at least 16 hours per week for a minimum of 6 months
  

Very nice-to-haves (hey, we have to dream):

• Experience with frameworks like LangChain, LangGraph, or similar orchestration tools

• Familiarity with FastAPI or other async Python web frameworks

• Exposure to vector databases (Milvus, Pinecone, Weaviate, etc.)

• Basic understanding of evaluation and monitoring for LLM applications

• You've deployed something to production, even if it was small