Staff AI Infra Architect: Scale the Intelligence Layer

Aimlroles; United States; Remote; USD 150,000 - 210,000; Full time

Job summary

Cassi is seeking a Staff- or Principal-level AI Infrastructure Engineer to own the intelligence layer end to end—the agent runtime, multi-provider inference infra, and surfaces used by customers.

You will audit existing architecture, drive migrations, and ensure scale, reliability, and cost efficiency across streaming, retrieval, and evaluation components. This is a production-focused, productization role at scale.

Qualifications

• 8+ years of professional backend engineering.
• Roughly 3–5 years working with LLMs.
• Production experience, not demos.
• TypeScript mastery in end-to-end systems.
• Experience with agent orchestration and tool-calling.
• Strong understanding of distributed systems and architecture.
• Experience with NoSQL and relational databases.
• Ability to ship features end-to-end in a small company.

Responsibilities

• Own end-to-end intelligence layer including agent runtime.
• Audit current design and plan scalable migrations.
• Own graph topology, tool calls, and multi-turn state.
• Manage streaming, failover, and provider routing in inference.
• Develop embedding and vector retrieval across corpora.
• Develop evaluation, observability, and unit economics.

Skills

Backend engineering

LLMs

TypeScript

Agent tooling

Inference infra

Streaming

Retrieval

Evaluation

Security

Architecture judgment

Education

Bachelor's degree in CS/related field

Tools

LangGraph

LangChain

Vercel AI SDK

DynamoDB

PostgreSQL

Job description

Cassi is seeking a Staff- or Principal-level AI Infrastructure Engineer to own the intelligence layer end to end—the agent runtime, multi-provider inference infra, and surfaces used by customers.

You will audit existing architecture, drive migrations, and ensure scale, reliability, and cost efficiency across streaming, retrieval, and evaluation components. This is a production-focused, productization role at scale.