remoteonsite
AI Engineer - SaaS Platform - Marrina Decisions
AI Engineer
AI Engineer responsible for maintaining and scaling LLM/VLM‑powered microservices on a SaaS platform, optimizing pipelines, managing async queues, and building feedback loops to continuously improve AI‑generated content and compliance scoring.
About the role
Key Responsibilities
- Maintain and optimize Flask/FastAPI microservices that power LLM and VLM content generation, compliance scoring, and campaign testing.
- Scale and monitor Dramatiq queues for asynchronous AI workflows, ensuring reliable orchestration of campaign generation pipelines.
- Deploy, observe, and troubleshoot Uvicorn/Gunicorn‑based services to achieve high uptime and low latency in production.
- Implement feedback mechanisms that capture user and system signals to continuously refine model outputs.
- Collaborate with data scientists and product teams to integrate new model versions and feature enhancements.
Requirements
- Strong proficiency in Python and experience building production‑grade Flask or FastAPI services.
- Hands‑on experience with large language models (LLM) and vision‑language models (VLM) in a SaaS environment.
- Familiarity with asynchronous task processing using Dramatiq or similar queue systems.
- Knowledge of deploying and monitoring Uvicorn/Gunicorn servers in cloud or containerized settings.
- Ability to design and implement feedback loops for model performance improvement.
Skills
pythonflaskfastapillm