Research Engineer (Kineto) (m/f/d)

JetBrains GmbH; München; Vor Ort; EUR 85.000 - 120.000; Vollzeit

Zusammenfassung

Kineto is seeking a Research Engineer to ship AI product features that blend engineering and ML. You will build and refine agent workflows, optimize latency and cost, and keep systems reliable as you track the evolving AI landscape for shipped products.

You will design self-improving loops that learn from feedback, run experiments, and drive product improvements end to end. This role suits someone who has shipped real systems and can own outcomes in a fast-moving startup environment.

Qualifikationen

• Shipped real systems, not just demos or prototypes.
• Experience building software with LLMs (structured outputs, tool use, reliability work).
• Built agents (multi-step workflows that call tools, keep states, and handle failures).
• Set up evals to track quality (tests, regression checks, eval harnesses).
• Comfortable moving fast, iterating, and owning outcomes end to end.

Aufgaben

• Ship AI product features that combine engineering and ML.
• Build and improve agent workflows (tooling, retries, guardrails, fallbacks).
• Improve the latency, cost, and reliability of our systems.
• Track what's new in the AI space (agent teams, methods, tooling) and turn it into shipped products.
• Build self-improving loops: systems that run experiments, learn from feedback and evals, and improve the product over time.

Kenntnisse

Shipped systems

ML fundamentals

Agent workflows

Experimentation & evals

End-to-end ownership

Ausbildung

Bachelor's in Computer Science or related

Tools

Python

PyTorch/TF

LLM tooling

Jobbeschreibung

Kineto (kineto.app) is an AI startup backed by JetBrains. We're building the next creator stack - tools that help creators talk to their audience in new ways, turn ideas into content, and build sustainable businesses - powered by LLMs and agents.

We're now looking for a Research Engineer with a strong grounding in engineering or ML who wants to turn ideas into shipped products.

How we work

We move fast and iterate. We ship early, measure results, and improve quickly. If something works, we double down. If it doesn't, we cut it. Delivery and measurement are part of the culture, and everyone owns outcomes end to end.

In this role, you will:

• Ship AI product features that combine engineering and ML.
• Build and improve agent workflows (tooling, retries, guardrails, fallbacks).
• Improve the latency, cost, and reliability of our systems.
• Track what's new in the AI space (agent teams, methods, tooling) and turn it into shipped products.
• Build self-improving loops: systems that run experiments, learn from feedback and evals, and improve the product over time.

We'd be happy to have you on our team if you:

• Have shipped real systems, not just demos or prototypes.
• Have built software with LLMs (structured outputs/function calling, tool use, reliability work).
• Have built agents (multi-step workflows that call tools, keep states, and handle failures).
• Have set up evals to track quality (tests, regression checks, eval harnesses).
• Feel comfortable moving fast, iterating, and owning outcomes end to end.