Research Scientist (AI Behaviours)

WhitecircleAI Safety company

Paris, FranceSenior

Romain Huet

Dirk Kingma

Guillaume Lample

Thomas Wolf

Olivier Pomel

François Chollet

Data & AI

Apply on site Bookmark

About the role

TL;DR

Research how LLM agents fail in real-world scenarios and build understanding of misalignment and unsafe behaviours.

View original posting →

Required skills

PythonTensorFlowPyTorchscikit-learnOpenAI APIAnthropic APILangChainHugging FaceLLMsPandasNumPyAirflowPostgreSQLMySQLMongoDB

Nice-to-have skills

A/B TestingProduct AnalyticsFigmaVisual DesignMotion DesignSEOCopywritingM&AContract ManagementCompliance Management

Domain expertise

Benefits & perks

Paid time off, Comprehensive medical insurance, Covered subscriptions for AI agents and IDEs, Team off-sites

Tech stack

PythonTensorFlowPyTorchscikit-learnOpenAI APIAnthropic APILangChainHugging FaceLLMsPandasNumPyAirflowPostgreSQLMySQLMongoDBRedisElasticsearchAWSGoogle CloudCI/CDDockerKubernetesGitHub ActionsGitLinuxJiraConfluencePostmanREST APISlack