Andreas Hermann
Andreas Hermann

AI Safety Researcher · ML Engineer

Andreas Hermann

I'm an ML engineer and researcher moving into AI safety full-time. I did a PhD, then spent three years taking machine-learning systems into production. Now I work on two things: keeping open-weight models safe after they're released, and the misalignment that shows up when you wire models together into agents.

Now

I'm doing independent AI safety research, funded by a transition grant that runs through April 2027. From September to December I'm also a research fellow at the Singapore AI Safety Hub (SASH). My project there asks which deployment conditions make open-weight AI agents take unsanctioned actions. It uses propensity evaluations with honeypots in a simulated cloud-infrastructure environment, and compares how conditions such as oversight and irrelevant noise shift the misbehavior rate across models. I'm also finishing an inoculation project with Safe AI Germany.

I'm open to research collaborations, especially where the question is whether safety holds up in deployment.

as of October 2026

The bet

I spent years keeping machine-learning systems running in production. That work shaped how I think about safety. I've watched which safety properties hold up once a model leaves the lab, and which ones quietly fall apart, and I want to work on that gap.

There's real research behind it too. I did a PhD, published 20+ peer-reviewed papers (490+ citations), and spent the last year properly upskilling through ARENA, the AI Alignment Research Fellowship, and BlueDot. Read the longer version →

Career

  1. 2026

    AI Safety Research Fellow · Singapore AI Safety Hub (SASH)

    Propensity evaluations of open-weight AI agents: which deployment conditions, such as oversight, permissions and irrelevant noise, make them take unsanctioned actions.

  2. 2026 – present

    Independent AI Safety Researcher

    Full-time research on open-weight model safety, compositional misalignment, and interpretability, backed by a BlueDot Impact transition grant.

  3. 2026

    AI Safety Research Fellow · Safe AI Germany (SAIGE)

    Inoculation against model poisoning: training a model so its safety properties can't just be fine-tuned away later.

  4. 2026

    Facilitator · BlueDot Impact

    Teaching technical AI safety cohorts and contributing to the alignment curriculum.

  5. 2023 – 2026

    ML / AI Engineering Lead

    Three years taking machine-learning systems into production at scale, across classification, data extraction, and data-quality models.

  6. PhD

    Doctorate · University of Münster

    Information systems. REACH Award for the dissertation.

Research focus

Open-weight model safety

What actually survives fine-tuning, quantization, and open release, once the deployment-time guardrails are gone.

Compositional misalignment

Why alignment tested on single models fails to compose once you chain them into multi-agent orchestrations, tool chains, and memory-augmented agents.

Interpretability & evaluation

The measurement tooling for safety: probes, evals, and mechanistic analysis that hold up in a real deployment.

See the full research agenda →

Selected publications

All publications & citations →

Recent writing

All writing →

Reading list

I curate and annotate papers, courses, and tools for AI safety researchers and engineers crossing over from adjacent fields.

Browse the reading list →