Staff AI Engineer - Architecture skills

shifttechnology; United States; Hybrid; USD 170,000 - 250,000; Full time

Benefits offered by this job

Flexible remote and hybrid options

Learning & development sessions

Generous PTO and holidays

Mental health benefits

Volunteering days

Job summary

Shift Technology seeks a senior, hands-on engineer to shape the architecture of agentic AI systems for an insurance-focused SaaS platform. You will design, prototype, and ship next-generation AI capabilities while mentoring teams and driving technical strategy, yet remaining deeply involved in coding.

You will build production-ready agentic workflows, convert prototypes into low-latency microservices on a major cloud platform, and integrate advanced RAG patterns, vector pipelines, and structured

Qualifications

• Hands-on experience with agentic frameworks (LangGraph, LlamaIndex, OpenAI Agents SDK, Microsoft Agent Framework) or similar orchestration.
• Experience with Model Context Protocol, tool/function calling, structured LLM outputs, and complex RAG pipelines.
• LLMOps tooling experience: evaluation, tracing, prompt engineering, and governance in production.
• Strong Python and OO programming for AI ecosystem projects.
• Backend microservices and REST/OpenAPI endpoints on enterprise cloud (Azure preferred).
• Familiarity with vector/relational databases, event-driven patterns, Docker/Kubernetes, CI/CD, and infra-as-code.

Responsibilities

• Design and build production-grade agentic workflows, orchestration layers, and tool-calling integrations.
• Transition AI prototypes into low-latency microservices on a major cloud platform.
• Integrate Model Context Protocol, RAG, vector pipelines, and structured LLM outputs into production.
• Establish LLMOps: experiment tracking, evaluation, latency optimization, token governance, observability.
• Mentor engineers, lead design reviews, set coding standards, and raise AI engineering capability.
• Benchmark new models/tools and translate breakthroughs into product value.

Skills

Python

Agentic frameworks

LLM frameworks

OO programming

Backend microservices

REST/OpenAPI

Cloud deployment

Tools

LangGraph

LlamaIndex

OpenAI Agents SDK

Microsoft Agent Framework

Docker

Kubernetes

Azure

Job description

Role overview

A senior, hands-on engineering role focused on shaping the architecture of agentic AI systems for an insurance-focused SaaS platform. The position sits at the intersection of applied LLM research and production-grade software engineering, with responsibility for designing, prototyping, and shipping next-generation AI capabilities. It is intended for a builder who wants to drive technical strategy while remaining deeply involved in coding and mentoring.

Responsibilities

• Design, prototype, and build production-ready agentic workflows, custom orchestration layers, and tool-calling integrations using modern AI frameworks.
• Transition AI prototypes into robust, low-latency microservices on a major cloud platform, ensuring tight integration with the surrounding tech stack.
• Integrate emerging paradigms such as Model Context Protocol, advanced retrieval-augmented generation patterns, vector pipelines, and structured LLM outputs into the production stack.
• Establish LLMOps practices covering experiment tracking, evaluation, prompt engineering, latency optimization, token budget governance, and model observability.
• Mentor engineering squads, lead technical design reviews, set coding standards, and raise the team's overall AI engineering capability.
• Continuously benchmark new models, tools, and architectures, translating relevant breakthroughs into practical product value.

Requirements

• Proven, hands-on experience building with agentic frameworks such as LangGraph, LlamaIndex, OpenAI Agents SDK, Microsoft Agent Framework, or custom orchestration solutions.
• Practical expertise with Model Context Protocol, tool and function calling, structured LLM outputs, and complex RAG pipelines including vector retrieval, hybrid search, and embeddings.
• Experience with LLMOps tooling, including evaluation frameworks, tracing, prompt engineering, and compute or token governance in production.
• Strong proficiency in Python and object-oriented programming for AI ecosystem work.
• Track record building and scaling backend microservices and REST or OpenAPI endpoints on enterprise cloud infrastructure, preferably Azure.
• Familiarity with vector and relational databases, event-driven patterns, containerization with Docker and Kubernetes, CI/CD, and infrastructure-as-code practices.

Nice to have

• Comfort reading research papers and extracting practical product value from emerging models and libraries.
• Demonstrated ability to upskill peers through pair programming, code reviews, and pragmatic design discussions.

Benefits and work setup

• Flexible remote and hybrid working options.
• Competitive base salary plus a variable component tied to personal and company performance.
• Multiple learning and development opportunities, including dedicated half-day learning sessions each month.
• Generous paid time off and paid holidays.
• Mental health benefits and paid volunteering days each year.
• Additional country-specific benefits may apply.