Master’s student in Computer Science, Saarland University.
Junior AI safety researcher.

juliuskamp2000 [at] gmail [dot] com · GitHub · LinkedIn · Saarbrücken, Germany

Education

Saarland University — M.Sc. Computer Science
2024 – present

Master’s thesis (in progress): Automated Debugger for AI agents.
Exploratory Data Analysis group (Prof. Jilles Vreeken) at CISPA, advisor: Nils Walter.

Saarland University — B.Sc. Data Science and Artificial Intelligence
2020 – 2024

Bachelor’s thesis: Discovering Fully Oriented Causal Networks from Discrete Observational Data
Exploratory Data Analysis group (Prof. Jilles Vreeken) at CISPA, advisor: Osman Mian.

Experience

Research Assistant — CISPA Helmholtz Center for Information Security
Dec 2024 – Nov 2025

Exploratory Data Analysis group.
Research on mechanistic interpretability in large language models.

Research Assistant — Saarland University
Oct 2023 – Oct 2024

Modeling and Simulation group.
Built and maintained ProcGrid Traffic Gym (PGTG), a Gymnasium-compatible reinforcement learning environmen. PyPI · Docs · GitHub

Web Developer — Brickmakers GmbH, Koblenz
May 2020 – Sep 2020

AI Safety

ARBOx4 (Alignment Research Bootcamp Oxford) — participant
Jun – Jul 2026

Two-week full-time intensive at Trajan House, Oxford. Compressed ARENA syllabus.
Capstone: replicated and extended the Intentional Control of Internal States experiment from Anthropic’s introspection paper on Gemma 3 27B. Write-up · Code

AI Safety Saarland (AISS) Advanced Fellowship — co-organizer
May – Jul 2026

Ten-week program on evaluations and interpretability. Ran the application process, speaker outreach, and some of the weekly sessions.

BlueDot Technical AI Safety Project Sprint — participant
May – Jun 2026

Scoped and ran a project on Natural Language Autoencoder–based measurement of evaluation awareness.

BlueDot Technical AI Safety course — participant
May 2026
AISS Interdisciplinary Research Incubator — participant
Nov 2025 – Feb 2026

Project on finding subspace circuits in transformers, applying the Subspace Partition method to toy transformer models.

Skills

  • ML / interpretability: PyTorch, TransformerLens, mechanistic interpretability (probes, activation steering, SAEs, circuit analysis), evaluations, reinforcement learning
  • Engineering: Python, Git, HuggingFace transformers, remote GPU workflows
  • Also familiar with: C, C++, HTML, JavaScript
  • Languages: German (native), English C1 (UNIcert III)