Senior AI R&D Engineer Model Training (f/m/d)

present of AI; Heidelberg; Ibrido; EUR 90.000 - 140.000; Tempo pieno

Descrizione del lavoro

Aleph Alpha in Heidelberg seeks a senior engineer to work across the full training stack—from cluster scheduling to evaluation throughput—and to improve code quality through testing and design. The role favors a software engineering mindset, prioritization, and the ability to unblock a team in a fast-moving research environment.

Collaboration through mob programming is encouraged, and the engineer will help build maintainable infrastructure that speeds up experimentation.

Competenze

• Experience designing robust training pipelines for large-scale AI systems.
• Ability to debug complex distributed systems in a research environment.

Mansioni

• Design and maintain training pipelines across the full stack from scheduling to evaluation throughput.
• Improve code quality through testing and incremental refactoring.
• Help make architectural decisions to speed up experimentation and maintainable infra.

Conoscenze

Mob programming

Test driven development

CI/CD

Refactoring

Prioritization

Debugging

Software engineering mindset

Strumenti

Python

Kubernetes

Rust

Descrizione del lavoro

The role asks for a senior engineer who can work across a full training stack, from cluster scheduling to evaluation throughput, and who can improve code quality through testing and design. A strong candidate brings a software engineering mindset, with experience in low coupling, high cohesion, and incremental refactoring, and is comfortable with collaborative practices like mob programming. The ability to prioritize and unblock a team in a fast-moving research environment is central.

How it fits the AI era

AI changes the work by automating some routine coding and testing, but the core tasks of designing robust training pipelines, debugging complex systems, and making architectural decisions remain human. The role stays human in its focus on collaboration, prioritization, and the judgment needed to build maintainable infrastructure that speeds up experimentation. AI tools may assist with code generation, but the engineer still needs to ensure the code is well tested and easy to understand.

How it builds AI experience

Working at Aleph Alpha offers direct exposure to pre-training large language models like Kolibri, giving hands-on experience with the infrastructure and tooling that make such training possible. A candidate can build AI experience by deepening skills in test driven development, trunk based development, and continuous integration, which are essential for maintaining research codebases. Familiarity with Python, Rust, and Kubernetes would also help, as these are used in the training pipeline.

Worth showing

• Trunk based development
• Incremental refactoring
• Python and Kubernetes

This is our own read, written from the employer's posting and the company's Spotlight. The employer's posting is the source of truth for the role.

About Aleph Alpha

Per the company, Aleph Alpha builds sovereign large language models and AI solutions for European enterprises and public institutions.