Applied Scientist — Multi-Modal AI & Research

Mistral; Palo Alto (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Mistral AI is seeking Applied Scientists and Research Engineers to drive innovative research and collaborate with clients on complex projects.

You will develop SOTA models across modalities (text, image, speech); create novel methods; and work with cross-functional teams to deliver high-impact AI solutions.

This role rewards self-starters who write clean Python and contribute to large codebases, with a collaborative environment and opportunities to publish.

Qualifications

• Excellent communication skills suitable for cross-audience.
• Expertise in PyTorch or JAX.
• Strong Python coding skills.
• Ability to work independently with minimal guidance.
• Collaborative and eager to learn.
• Proven track record through projects or academia.

Responsibilities

• Run preprocessing, training, and evaluation of SOTA models across modalities.
• Develop and implement novel methods for multi-modal AI tasks.
• Collaborate with cross-functional teams including engineers and product
• Contribute to data collection, tooling, and deployment pipelines.
• Communicate research results and impact to clients and stakeholders.

Skills

PyTorch or JAX

Python programming

Communication skills

Independent / self-driven

Collaborative

Research experience

Education

PhD or MSc in Mathematics/Physics/ML/CS & Engineering

Job description

Mistral AI is seeking Applied Scientists and Research Engineers to drive innovative research and collaborate with clients on complex projects.

You will develop SOTA models across modalities (text, image, speech); create novel methods; and work with cross-functional teams to deliver high-impact AI solutions.

This role rewards self-starters who write clean Python and contribute to large codebases, with a collaborative environment and opportunities to publish.