Are you an ambitious student looking to build a career at the overlap of AI and manufacturing? Are you excited about the latest developments in GenAI and agentic systems? Do you want to learn how to build & deploy AI solutions & RAG pipelines that actually have a purpose, inside a sovereign EU manufacturing AI platform? Then this might be an opportunity you have been waiting for & we are looking forward to meeting you!

Project:

Calibras is developing an advanced Manufacturing AI system, tailored for automating critical industrial workflows, supporting domain specific data formats. With proprietary context engineering scaffolding, our models are capable of outperforming frontline models, at a lower price - with on-premise option, built with 100 % EU tech stack.

Main scope of the project is to build up on existing RAG system, and introduce a V2 of our feature extraction pipeline - supporting the next level of customer requirements: persistent & adaptive agentic memory, improved accuracy of extracted manufacturing features and increased performance of the LLMOps structure.

The project is envisioned as a student internship lasting 1-3 months - with the intention of promotion to a part-time student position.

Candidate profile & application:

• Strong understanding of RAG & agentic architectures, data pipelines for LLM systems (chunking strategies, embeddings, vector databases, feature engineering). Experience with LangChain & LangGraph. Domain knowledge of manufacturing is a huge plus.

• Hands-on experience with MLOps/LLMOps practices, including evaluation frameworks, regression testing, monitoring, and observability. Experience with deploying local LLMs – knowledge of Linux, GPU VRAM optimization, basic networking & agentic frameworks (like Hermes or OpenClaw).

• Good Python skills with production standards (testing, CI/CD, performance awareness).

• Structured mindset and willingness to learn. Genuine passion for engineering & good team spirit.

Application requirements: Short CV. Brief introduction / cover letter are optional.

We are looking forward to hearing from you!

Calibras