Metadata Intelligence & RAG Systems Internship - Open Day Edition

Descrizione del lavoro:

Career Category
Supply Chain

Job Description

(Limited Spots - Attendance by Selection Only)
Are you driven by applied machine learning, metadata intelligence, information retrieval, and cutting-edge LLM workflows?
This assignment is part of the Amgen Internship Open Day, where four internship opportunities will be presented directly by their hiring managers.

📅 Open Day Date: 16 April
🔒 Limited spots available
🎯 Participants will be selected by the hiring managers
📬 Final confirmations will be sent during the week of 6 April

⚠️ Important: Once you confirm your spot, attendance is mandatory - not attending means taking a seat from another motivated student.

About the Internship - Metadata Intelligence: Theory + Prototype
This 6-month final-year internship offers a rare opportunity to work on state-of-the-art metadata enrichment, semantic search, embeddings, and RAG (Retrieval-Augmented Generation) inside Amgen's production ecosystem.
You will translate academic techniques into practical, reproducible prototypes using Amgen's high-quality datasets and collaborate closely with analytics teams, engineers, and domain SMEs.
This assignment is ideal for students wanting to bridge academic research with industry-ready implementation.

Your Assignment
Over six months, you will design and deliver a full metadata intelligence workflow:
1. Research & Theory
* Conduct a literature review on modern metadata enrichment, semantic search, and RAG workflows
* Compare rule-based, Transformer/LLM-based, and hybrid extractors + retrievers
* Build hypotheses and design evaluation frameworks

2. Prototype Development
* Implement baseline extractors
* Build LLM prompt pipelines, embedding-based retrievers, and hybrid workflows
* Integrate a RAG-enabled helper tool/playbook
* Develop a metadata extractor + embedding + retrieval pipeline using Amgen datasets

3. Dynamic Data-Maturity & Health Reporting
* Build an interactive dashboard showing catalog completeness, data quality, pipeline freshness, KPI health
* Demonstrate how improved metadata + IR methods drive measurable productivity gains

4. Final Deliverables
* Annotated literature review
* Reproducible prototype code
* Evaluation suite with metrics & recommendations
* Integrated RAG demo
* Interactive data-maturity dashboard
* Handoff documentation + final stakeholder presentation

What You Will Learn
This internship provides hands-on experience in:
* End-to-end ML/NLP research and prototyping
* RAG workflows, embeddings, and metadata extraction
* Information retrieval evaluation (intrinsic + task-based metrics)
* Production-grade reproducibility and experimentation
* Stakeholder interaction within a regulated, high-impact data environment
* Presenting advanced technical concepts to non-technical audiences

You will finish with portfolio-quality code, reports, dashboards, and demos.

Skills & Profile We're Looking For
✅ Must-Haves
* Basic Python (scripts, pandas, small experiments)
* SQL fundamentals (SELECT, JOIN, GROUP BY, window functions)
* Experience working in notebooks (Jupyter/Colab)
* Problem-solving mindset and ability to debug
* Clear written & verbal communication

✅ Nice-to-Haves
* Exposure to ML/NLP concepts (classification, embeddings, transformers)
* Experience with git / basic version control
* Any prior work with vector search or embeddings
* Experience with Databricks, SQL, Power BI (preferred)

🎓 Internship Details
* Type: Final-year internship (Master's preferred)
* Duration: 6 months
* Start: September 2026
* Preferred Universities: Tilburg University, Erasmus University, TU Eindhoven, TU Delft
* Preferred Majors: Econometrics, Data Science, Business Analytics & OR, Software Engineering, Statistics, Applied Mathematics, ML/AI

🌱 What Amgen Offers
* Structured mentorship (1-2 hrs/week)
* SME access for annotation & evaluation
* Compute resources + production environment
* Stakeholder exposure and presentation opportunities
* A diverse, international, collaborative environment
* Hybrid working flexibility
* Gym, vitality programs, healthy cafeteria
* Possibility to apply for temporary or permanent roles

🎟️ About the Open Day - Important Information
* Spots for the Open Day are limited
* Participants will be selected personally by the hiring managers
* Final confirmations after selection will be sent during the week of 6 April
* If you confirm your attendance, you are required to show up - declining too late or not attending removes an opportunity from another student
* This DRP internship will be one of four internships presented during the Open Day