Own agent quality for Bliro's chat and voice agents, from prompting and context to model benchmarking and evals.
Coordinate and evaluate models rather than training them, working within a small team of specialists.

What You'll Do:

In 30 days, map agents end to end, review real user conversations, and ship an initial improvement.
In 60 days, build eval sets and metrics, benchmark models on quality, latency, and cost, and ship a measurable improvement.
In 90 days, establish a repeatable improvement model and set the roadmap for agent quality.

Who You Are:

You have evidence of shipping LLM agents to real users and proving they got better with data.
You have strong Python and can debug agent behavior across prompts, tools, context, and model choice.
You build eval pipelines with LLM judges and use statistics to tell real gains from noise.

Why Bliro:

We build the human interface for millions of field sellers with a culture of ownership and urgency.
We offer a strong base salary plus meaningful equity and a hybrid setup in Munich.

Bliro GmbH

Bliro builds an AI assistant that handles desk work for field sales teams, with over one million customer touchpoints documented. Backed by leading investors and trusted by German Mittelstand companies, it's a small, high-performing team that values ownership and continuous improvement.

Apply for This Position