Tech Stack
BigQueryGoogle Cloud PlatformKubernetesPythonTerraform
About the role
- Develop, refine, and enable the AI Agent to automatically address and resolve shopper requests for merchant clients
- Own feedback, observability, and continuous training of deployed solutions/models (HuggingFace, Vertex, Comet ML, multiple LLM providers, PromptLayer)
- Work on Generative Agents, Recommender systems, and Knowledge graphs
- Solve product-facing problems with classical NLP and/or with LLMs
- Build clear tech specs and implement domain-driven design solutions that scale
- Maintain services (FastAPI) and cron jobs (Kubeflow)
- Own development lifecycle (Terraform permissions, test coverage of 80%, CI/CD, k8s, Datadog, PagerDuty)
- Collaborate cross-functionally to provide Machine Learning as a service within Gorgias
- Maintain ML team data pipeline (BigQuery and DBT)
- Analyze post-release performance and identify improvements for next iterations
Requirements
- 5 years of proven track record in Machine Learning (NLP/HuggingFace/Math), software design (DDD) and developing & maintaining APIs and batch jobs (FastAPI/kubeflow/DBT)
- Familiarity with LLM life cycle, text-to-speech, speech-to-text, reinforcement learning and vision (plus)
- Proficiency in Python and familiarity with engineering tools and frameworks
- Familiarity with GCP (plus)
- Excellent written and verbal communication skills in english
- Masters degree in STEM (science, technology, engineering and mathematics) or a related field
- Ability/willingness to be based in or relocate to Paris and commit to hybrid work (see application form)
- Legal authorization to work in the job location (see application form)