Full-Stack AI Engineer

Responsibilities:
- Help design and build MARV's own AI infrastructure.
- Shape the architecture we own rather than just integrating third-party model APIs.
- Design and ship RAG pipelines over legal and financial documents including embeddings, retrieval, and chunking strategies.
- Build and orchestrate the agentic workflows that power MARV's core product experience on infrastructure we control.
- Own backend infrastructure end-to-end including APIs, vector databases, cloud deployment, and scaling.
- Build evaluation and guardrail systems to keep hallucination and error rates low in a domain where mistakes matter.
- Work directly with our founding team to shape product, architecture, and roadmap.

Qualifications:
- 3+ years of production software engineering experience with hands-on work shipping LLM-powered features.
- Genuine interest in building proprietary AI infrastructure rather than only integrating third-party model APIs.
- Real experience with RAG including vector databases, embeddings, and retrieval tuning.
- Strong backend fundamentals in Python and/or TypeScript.
- Experience with API design, cloud infra, and containerization.
- Comfort with agentic frameworks and a willingness to build custom orchestration.
- Exposure to open-source or self-hosted model deployment such as fine-tuning, inference optimization, and model serving.
- A bias toward shipping and iterating over perfecting in isolation.
- Experience in a regulated industry like legal, fintech, or healthcare is a bonus.

Location: Remote - Anywhere

Skills required for this job:

• API design
• Agentic AI systems
• Back-end development
• Cloud platforms
• Containerization
• Deployment strategies
• Embedding
• Retrieval-augmented generation (RAG)
• Software engineering
• Vector databases