Materials Science Expert - Engineering

Obsidian; Berlin; Vor Ort; EUR 70.000 - 110.000; Vollzeit

Zusammenfassung

Obsidian is seeking a senior professional to design domain-authentic tasks for evaluating AI models. You will author tasks in materials science or electrical/mechanical engineering domains, build data rooms and grading criteria, and run evaluations to iterate toward well-defined success criteria.

The role requires navigating ambiguity, explaining not only correct solutions but why incorrect ones fail, and collaborating with a project lead and domain experts to define what good looks like.

Qualifikationen

• Senior expertise in materials science — metals, ceramics, composites, characterization, additive manufacturing — or in electrical engineering (analog, RF, power) or mechanical engineering (design, CFD and fluids, heat transfer, pressure vessels, NDE, CAE).
• Comfortable navigating ambiguity — much of the work is still being defined, and shaping it is part of the role.
• Able to explain not just the right answer, but why a wrong answer is wrong.

Aufgaben

• Author domain-authentic tasks drawn from work you've actually done.
• Build the data room and grading criteria that make the task objectively scorable.
• Run the task against the model, analyse where it succeeds or fails, and iterate until it genuinely discriminates.
• Work with the project lead and other domain experts to settle what "good" looks like in your domain.

Kenntnisse

Materials science

Electrical engineering

Mechanical engineering

Explain reasoning

Jobbeschreibung

About the role

We build materials-science and engineering tasks that test how well an AI model does real expert work. You write a realistic task — a prompt, a data room, and a way to grade it — run it against the model, and tighten it until the model can no longer reason through it cleanly.

What you'll do

• Author domain-authentic tasks drawn from work you've actually done.
• Build the data room and grading criteria that make the task objectively scorable.
• Run the task against the model, analyse where it succeeds or fails, and iterate until it genuinely discriminates.
• Work with the project lead and other domain experts to settle what "good" looks like in your domain.

Who we're looking for

• Senior expertise in materials science — metals, ceramics, composites, characterization, additive manufacturing — or in electrical engineering (analog, RF, power) or mechanical engineering (design, CFD and fluids, heat transfer, pressure vessels, NDE, CAE).
• Comfortable navigating ambiguity — much of the work is still being defined, and shaping it is part of the role.
• Able to explain not just the right answer, but why a wrong answer is wrong.

Support

Daily onboarding sessions and multiple office hours every day.