AI Agent Systems Architect Responsibilities in AI DevelopmentArchitect, design, and actively write code for production-grade stateful multi-agent systems, custom state machines, and function-calling workflows using frameworks such as LangGraph, CrewAI, or native Python execution loops.
Lead full-cycle hands-on development: data preparation, tool integration via Model Context Protocol (MCP), structured output enforcement, context-window recovery/caching, containerization, and secure cloud deployment (Azure/AWS).
Implement MLOps/LLMOps evaluation pipelines (e.g., Langfuse, Ragas) for automated prompt regression tracking, hallucination mitigation, and human-in-the-loop (HITL) gating.
Apply Responsible AI practices and guardrails throughout solution delivery to ensure system reliability and safety.
Translate business and mission requirements into technical designs; prototype and iterate quickly with stakeholders using AI-accelerated workflows.
Contribute to Azumo's innovation roadmap by identifying research topics, publishing insights, and advancing our AI software development services portfolio.
RequirementsBasic Qualifications:- Bachelor's Degree in Computer Science, Data Science, or related field (Master's is a plus).
- 3+ years of experience developing and deploying ML, NLP, or Generative AI systems in production environments.
- Expert-level skills in Python and software engineering fundamentals (data structures, async programming, API design, testing, CI/CD, Git, containers).
- Hands-on experience with stateful AI agent frameworks and tooling: LangGraph, LangChain, CrewAI, MCP, and vector databases (Pinecone, LanceDB, Azure AI Search).
- Active experience utilizing modern AI-assisted coding tools (e.g., Claude Code, Cursor, GitHub Copilot) to accelerate development and codebase analysis.
- Proven cloud deployment experience (Azure preferred; AWS acceptable) using Docker and serverless/microservice architectures.
- Strong written and verbal communication skills to explain complex architectural concepts and trade-offs to diverse audiences.
- Professional English proficiency (C1).
Preferred Qualifications:- Experience building Human-in-the-Loop (HITL) workflows, automated evaluation suites, and deterministic fallback logic for LLMs.
- Contributions to research papers, open-source AI libraries, or active participation in the AI engineering community.
Benefits- Paid time off (PTO)
- U.S. Holidays
- Training
- Mentored career development
- Profit Sharing
- $US Remuneration
