Skip to content
Pegel

ML engineer jobs in Berlin with no German required

12 active roles

Machine learning engineering in Berlin is almost entirely an English-language specialisation. The talent pool that builds and ships ML systems is international by necessity, the research literature the work draws on is in English, and the companies in Berlin with serious ML engineering needs are precisely the ones that hired globally from the start. I collect language classification data across thousands of Berlin startup job descriptions, and ML engineering roles have the highest no-German rate of any technical discipline I track. That is not a marginal difference; it is the characteristic of the field here.

What this page shows: roles where someone is expected to put machine learning into production, own the model serving layer, build training pipelines, or maintain the evaluation infrastructure that keeps a model honest over time. This is not the same as a data scientist role, though the titles blur often enough that you should read the job description rather than rely on what it is called. The strongest listings are explicit about whether the work is modelling-first or engineering-first. Python and PyTorch are the common core. MLflow or a similar experiment tracker, a feature store, and Kubernetes-based serving infrastructure appear at companies that are past the prototyping stage.

Seniority at the no-German end of ML engineering skews senior. AI-native Berlin startups are often past Series A and have real production systems to maintain. They are not typically looking for someone who needs mentoring on the basics of model deployment. Mid-level roles exist, but the no-German cohort is smaller at that level than in web engineering, because fewer ML-native companies are at the stage where mid-level hiring makes sense.

Pegel pulls these from each company's public careers feed and refreshes daily.

Open roles

Questions

Why is ML engineering in Berlin almost entirely English-speaking?
The discipline is young enough that no single country dominates the pipeline of engineers who know it well. Berlin's AI companies hired from wherever they could find the right skills (Eastern Europe, Spain, the UK, the US, India) and the shared working language was English by default. That culture is self-reinforcing: an English-first ML team keeps posting English-first ML roles.
What distinguishes an ML engineering role from a data scientist role on this page?
I do not separate them perfectly at the title level, but the strongest signal is whether the role expects production ownership. An ML engineering role should own the path from training to serving, including latency and reliability, and model performance in production. A data scientist role tends to stop at the research or analysis stage. When the title is ambiguous, the requirements section usually clarifies which kind of ownership is expected.
Do Berlin AI startups offer equity to ML engineers as a meaningful part of compensation?
Some do, more than in other sectors. Berlin startups across the board have become more serious about equity as a compensation component, and AI companies competing for a small pool of production ML engineers are among the more aggressive at offering it. Whether the equity is meaningful depends on the stage, the valuation, and the structure of the grants, which the job description will not tell you. You will need to ask.
DAILY DIGEST

New Berlin startup roles in your inbox, every morning, Monday to Saturday. Pulled straight from company career pages, filtered to exactly what you ask for. Your current filters are saved with the subscription.

Double opt-in, one-click unsubscribe, no tracking. See privacy for the details.

Related searches