Senior Artificial Intelligence Agent Engineer

About the Role

We are seeking a Senior AI Agent Engineer to design, develop, and deploy production-quality AI agent systems. This role requires deep expertise beyond simple LLM prompt connections, focusing on architecting robust single and multi-agent systems, managing workflows, tool integration, memory, retrieval-augmented generation (RAG), automation, safety, evaluation, and observability.

The systems you build will operate reliably in real-world scenarios, interfacing with external tools and APIs, executing complex multi-step tasks, and scaling across varied customers and environments.

Key Responsibilities

• Design and implement architectures for single-agent and multi-agent systems.
• Create dependable multi-step workflows and execution patterns for agents.
• Develop orchestration frameworks such as planner-executor, supervisor-worker, routing, delegation, and hierarchical agents.
• Develop reusable agent features including tools, memory systems, policies, and safety guardrails.
• Build systems balancing autonomy, dependability, cost effectiveness, and operational performance.
• Construct stateful workflows using frameworks like LangGraph, LangChain, or similar.
• Implement structured outputs and deterministic validations for agent decisions and tool invocations.
• Develop event-driven workflows triggered by various system or user events.
• Apply asynchronous processing technology such as RabbitMQ, Kafka, BullMQ, or equivalents.
• Engineer resilient execution systems with mechanisms for retries, idempotency, checkpointing, timeouts, and error recovery.
• Create custom tools enabling agents to interact securely with external systems.
• Integrate agents with APIs, databases, applications, and authorized data sources while enforcing access controls.
• Implement Model Context Protocol (MCP) integrations where necessary.
• Ensure agent actions are executed safely without unauthorized system or data access.
• Design and maintain production-level RAG pipelines involving embeddings, vector search, filtering, and reranking.
• Implement short-term and long-term memory methods along with context management strategies such as retrieval, summarization, and persistence.
• Protect agents from prompt injections and inappropriate tool usage.
• Develop permission systems, human approval checkpoints, and safe failure modes for sensitive operations.
• Build automated evaluation tools focusing on task completion, safety, reliability, and decision quality.
• Establish observability frameworks to monitor agent execution including latency, failures, token consumption, and operational cost.
• Implement usage metering and optimize model selection, token usage, latency, and cost efficiency.

Qualifications and Desired Experience

• Proven track record in building and rolling out agentic AI systems in production settings.
• Expertise in agent orchestration, multi-step executions, structured tool calls, and workflow design.
• Hands-on experience with LangGraph, LangChain, or similar orchestration frameworks.
• Proficiency in developing custom tools and integrating with external APIs securely.
• Experience with messaging technologies like RabbitMQ, Kafka, or BullMQ.
• Knowledge of RAG techniques, vector databases, embeddings, and memory management for AI agents.
• Solid understanding of implementing safety guardrails, approval workflows, and permission controls.
• Familiarity with agent evaluation metrics, tracing, and observability tools.
• Strong programming skills in Python and/or TypeScript/Node.js.
• Adeptness in software engineering concepts including RESTful APIs, distributed and asynchronous systems, testing, and system architecture.
• Insight into LLM performance characteristics such as latency, token economics, and cost optimization.
• Experience managing AI systems in live production environments.

Preferred Experience

• Building AI platforms with multi-agent architectures.
• Developing event-driven systems and distributed infrastructures.
• Utilizing MCP and tool ecosystems within AI agents.
• Constructing evaluation frameworks for agent reliability and performance.
• Contributions to open-source AI or agent-related projects.
• Comprehensive understanding of LLM limitations and advanced agent architectures.

Additional Information

This position requires a focus on building production-grade AI systems from end to end and is not suitable for those whose experience is limited to prompt engineering, chatbot design, or prototype development. The role is fully remote and accessible to applicants worldwide.

Work Location: Anywhere worldwide (Remote)

Employment Type: Full-time