Post-Training Research Scientist, Speech AI

nyra health; Wien; Hybrid; EUR 75.000 - 115.000; Vollzeit

Benefits dieser Stelle

Phantom Stock Options

Hybrid working model

Beautiful Vienna office

Zusammenfassung

nyra health is building software that supports clinics, therapists, and patients in neurorehabilitation. nyra labs, the research arm, develops post-training methods and datasets from real clinical data to make speech models safer and more useful in practice.

As a Researcher in Post Training, you’ll take hypotheses from idea to experiments, evaluating outcomes and releasing robust results that advance speech AI for underserved populations. Vienna-based hybrid work is available.

Qualifikationen

• Post-training expertise with fine-tuning, preference optimization, RLHF, distillation or related methods.
• Deep learning experience with PyTorch and modern model-training workflows.
• Design evaluations and diagnose complex model behavior across data, training, and inference.
• ML engineering ability to write production-quality Python and debug distributed training systems.
• Research rigor with publications, open-source work, or independent research.
• Advanced degree or equivalent in ML, speech processing, NLP, or related field.
• AI-native workflow using modern coding and research agents to move faster with scientific rigor.

Aufgaben

• Develop and test post-training methods including supervised fine-tuning, preference optimization, distillation, feedback-driven learning, and reinforcement learning where useful.
• Create high-quality post-training datasets through curation, annotation, synthetic data generation, and model-assisted data improvement.
• Improve model behavior: instruction following, verbatim transcription, uncertainty handling, long-form consistency, multilingual performance, and resistance to hallucinations.
• Build benchmarks and failure taxonomies that reveal whether models are genuinely improving.
• Implement, debug, and scale training and evaluation pipelines with a focus on reproducibility.
• Collaborate with research and product teams to adapt foundation models to specific speech and clinical use cases.
• Contribute to publications, model releases, datasets, and technical reports.

Kenntnisse

Post-training methods

PyTorch

Evaluation mindset

ML engineering

Research rigor

AI-native workflow

Collaboration

Ausbildung

MSc/PhD or equivalent in ML, speech/NLP

Tools

Python

Distributed training

Jobbeschreibung

nyra health is building software that supports clinics, therapists, and patients in neurorehabilitation. nyra labs, the research arm, develops post-training methods and datasets from real clinical data to make speech models safer and more useful in practice.

As a Researcher in Post Training, you’ll take hypotheses from idea to experiments, evaluating outcomes and releasing robust results that advance speech AI for underserved populations. Vienna-based hybrid work is available.