Jobs

AI Engineering Internship (RAG / LLMs / Python / R)

On our Jobs page, we publish selected open positions at RPACT for people who are excited about contributing to innovative software and applied statistical computing in a highly specialized domain.

We are currently looking for an AI Engineering Intern to help us explore and build the next generation of AI-assisted workflows for clinical trial design.

Build an AI assistant for clinical trial design (R + LLMs)

About the Project

We are developing a next-generation AI system that helps statisticians design and analyze clinical trials using natural language, while ensuring that all results are computed deterministically and reproducibly in R.

Unlike typical AI coding tools, this system:

  • uses real statistical computation (no hallucinated results)
  • integrates with the R package rpact for innovative clinical trial design
  • combines LLMs, retrieval (RAG), and structured execution
  • is designed for enterprise and regulated environments

You will play a key role in building the first MVP of this system, creating the foundation for future refinement and expansion.

What You Will Work On

Your main task will be to build an internal AI system that understands and assists with rpact code and documentation.

Concretely, you will:

  • build a retrieval-augmented generation (RAG) system over:
    • documentation
    • source code
    • tests
    • validation material
  • design and implement:
    • document parsing and preprocessing pipelines
    • intelligent chunking strategies (functions, tests, sections)
    • embedding and vector database setup (e.g. Chroma / FAISS)
  • develop a backend service for retrieval, orchestration, and LLM integration (e.g. Plumber, FastAPI, or similar)
  • connect the system to an LLM to enable:
    • natural language queries
    • context-aware code suggestions
  • build a simple chat interface to interact with the system
  • evaluate and improve retrieval quality

Depending on progress and interest, you may also:

  • prototype structured tool-calling (JSON → R execution)
  • generate reproducible R scripts automatically
  • explore integration with developer tools (e.g. GitHub Copilot)

What You Will Learn

  • how to build real-world AI systems beyond simple chatbots
  • retrieval-augmented generation (RAG) in practice
  • working with LLMs in structured, reliable workflows
  • designing systems for correctness and reproducibility
  • applying AI in a high-impact domain (clinical trials)

Requirements

Must-have

  • computer science background (Master’s level or higher)
  • strong programming skills in Python
  • familiarity with:
    • APIs / REST
    • JSON / data structures
    • Git / GitHub
    • R or willingness to work with R
  • ability to work independently
  • laser focus on getting difficult and challenging tasks done

Nice to have (not required)

Experience with:

  • LLMs / embeddings / RAG and the surrounding tooling ecosystem (frameworks and libraries are flexible)
  • NLP or machine learning
  • Plumber, FastAPI, or similar frameworks
  • Docker
  • basic knowledge of R, statistics, or scientific computing is beneficial, since statistical computation in the system is performed in R

What We Offer

  • a high-impact AI project with real-world application in a well-defined MVP
  • flexible working setup (fully remote)
  • close collaboration with experienced statisticians and developers
  • opportunity to contribute to a future product used in pharma and biotech
  • competitive compensation in the range of €1,000–€3,000 per month, depending on experience and location
  • potential for a subsequent permanent contract, depending on performance and motivation

Logistics

  • Location: EU
  • Duration: 3–6 months
  • Expected commitment: part-time (approximately 10–20 hours/week equivalent)
  • We focus on deliverables and milestones rather than fixed working hours.
  • The role is designed to be compatible with ongoing academic studies.
  • Start date: flexible

About Us

We develop statistical software for clinical trial design, including the R package rpact, which is widely used in industry and academia. Our focus is on innovative, reliable, and reproducible statistical methods. See our Company page for more details.

How to Apply

Please send the following to :

  • short CV including academic grades
  • short description of relevant experience (projects, coursework, GitHub)
  • optional: your own example project related to AI, NLP, or data science

We are particularly interested in candidates who enjoy building practical systems and exploring new technologies — curiosity and initiative matter more than perfect experience.