AI Safety & Alignment Research Scientist

Story Terrace Inc.; Paris; On-site; EUR 70,000 - 110,000; Full time

Job summary

Lexsi Labs in France is seeking a research scientist to push forward AI explainability, safety, and alignment at scale. You will tackle large-scale industry problems and collaborate with ML researchers and engineers in a startup‑like environment.

The role requires a strong academic background in ML/DL/RL, a Master or PhD in CS or Mathematics, publications or open‑source contributions, and hands‑on experience with TensorFlow and PyTorch; cloud/on‑prem deployment, and ML explainability methods.

Qualifications

• Solid academic background in ML/DL/RL concepts.
• Master or PhD in CS or Mathematics is required.
• Should have published peer‑reviewed papers or contributed to opensource tools.
• Hands‑on experience with DL frameworks like TensorFlow and PyTorch.
• Experience with diverse data types (text, tabular, images).
• Cloud and/or on‑prem deployment experience.
• Strong fundamentals in MLOps and productionising ML models.
• Experience with ML explainability methods (LRP, SHAP, LIME, IG, CEM).
• 2+ years of hands‑on DL/ML experience.
• Hands‑on experience with Transformer models, GANs, DL.

Responsibilities

• Work on AI explainability, safety, and alignment.
• Flexible in specialization areas within ML/DL.
• Create new techniques around ML Observability & Alignment.
• Collaborate with MLEs and SDE to roll out features and ensure quality.
• Create and maintain technical and product documentation.
• Publish papers in open forums like arxiv and present in industry forums like ICLR NeurIPS.

Skills

TensorFlow

PyTorch

ML/DL fundamentals

Transformer models

GANs

MLOps basics

Explainability methods

Education

Master's or PhD in CS or Mathematics

Job description

Lexsi Labs in France is seeking a research scientist to push forward AI explainability, safety, and alignment at scale. You will tackle large-scale industry problems and collaborate with ML researchers and engineers in a startup‑like environment.

The role requires a strong academic background in ML/DL/RL, a Master or PhD in CS or Mathematics, publications or open‑source contributions, and hands‑on experience with TensorFlow and PyTorch; cloud/on‑prem deployment, and ML explainability methods.