Artificial Intelligence Engineer

Job Type: Full-Time, RemoteLocation: Anywhere in the WorldExperience Level: SeniorAbout The Role;We are looking for a Senior AI Agent Engineer to design, build, and deploy production-grade AI agent systems.This is an engineering role for someone who understands that building reliable AI agents goes far beyond connecting an LLM to a prompt. You will be responsible for architecting single-agent and multi-agent systems, stateful workflows, tool execution, memory, RAG, automation, safety, evaluation, and observability.You will work on systems designed to operate reliably in real-world environments, interact with external tools and APIs, execute multi-step tasks, and scale across customers and use cases.The ideal candidate is an experienced software engineer who has actually shipped agentic AI systems into production, not someone whose experience is limited to AI prototypes, chatbots, or prompt engineering.Key Responsibilities;AI Agent Architecture:Architect and implement single-agent and multi-agent systems.Design reliable multi-step agent workflows and execution patterns.Build orchestration architectures including planner-executor, supervisor-worker, routing, delegation, and hierarchical agents.Design reusable agent capabilities including tools, skills, memory, policies, and guardrails.Build systems that balance autonomy, reliability, cost, and performance.Agent Workflows & Orchestrations:Build stateful workflows using LangGraph, LangChain, or equivalent frameworks.Implement structured outputs and deterministic validation for agent decisions and tool calls.Design event-driven workflows triggered by user requests, system events, webhooks, and scheduled jobs.Implement asynchronous processing using RabbitMQ, Kafka, BullMQ, or equivalent technologies.Build robust execution systems with retries, idempotency, checkpointing, timeouts, and failure recovery.Tools, API's & Integrations:Develop custom tools that allow agents to interact with external systems.Integrate agents securely with APIs, applications, databases, and approved data sources.Implement Model Context Protocol (MCP) integrations where appropriate.Build permission-controlled tool execution and access mechanisms.Ensure agents can safely execute actions without unauthorized access to systems or data.RAG, Memory & Context:Design and implement production-grade RAG pipelines.Work with embeddings, vector search, hybrid retrieval, metadata filtering, and reranking.Design short-term and long-term agent memory.Develop strategies for context management, retrieval, summarization, and persistence.Optimize retrieval quality, relevance, latency, and cost.Safety & Guardrails:Protect AI agents against prompt injection and malicious or unintended tool usage.Implement permission systems and guardrails around sensitive actions.Build human approval checkpoints for high-impact or sensitive operations.Prevent unauthorized data access and unintended agent actions.Design systems that fail safely when an agent encounters uncertainty or unexpected conditions.Evaluation & Observability:Build automated evaluations for:Task completionGroundednessTool selectionSafetyReliabilityAgent decision qualityImplement tracing and observability across agent execution.Monitor tool calls, latency, failures, token consumption, and operational costs.Identify failure patterns and continuously improve agent performance.Usage & Cost Optimization:Implement usage metering across customers, agents, models, and tasks.Monitor LLM usage and infrastructure costs.Optimize token consumption, model selection, latency, and execution efficiency.Build systems capable of scaling without uncontrolled AI or infrastructure costs.Required Qualifications;Proven experience building and deploying agentic AI systems in production.Strong understanding of agent orchestration, tool calling, structured generation, and multi-step execution.Hands-on experience with LangGraph, LangChain, or an equivalent agent orchestration framework.Experience developing custom tools and integrating external APIs.Experience with RabbitMQ, Kafka, BullMQ, or equivalent messaging technologies.Practical experience with RAG, embeddings, vector search, and agent memory.Experience implementing guardrails, approval workflows, and permission-controlled tool execution.Experience with agent evaluation, tracing, and observability.Strong proficiency in Python, TypeScript/Node.js, or both.Strong software engineering fundamentals, including APIs, distributed systems, asynchronous processing, testing, and system design.Understanding of LLM performance, latency, token economics, and usage optimization.Experience deploying and operating AI systems in production environments.What We're Looking For;We're looking for an engineer who can answer "How do we make this reliable in production?", not just "How do we make the model do this?"You should be comfortable working across:LLMs → Agents → Tools → APIs → Data → Workflows → Infrastructure → Evaluation → SecurityYou should also be comfortable making engineering trade-offs around reliability, latency, cost, scalability, and model performance.Strong Candidates Will Have Experience With;Production agentic AI platformsMulti-agent architecturesLangGraph / LangChain or equivalent frameworksTool calling and function executionRAG and vector databasesAgent memory and context managementEvent-driven architecturesMessage queues and asynchronous systemsMCPAgent evaluation and observabilityAI security and guardrailsLLM cost and performance optimizationWhat Would Make You Stand Out;Experience building AI agents that autonomously execute real business workflows.Experience designing multi-agent systems at scale.Experience with MCP and tool ecosystems.Experience building AI infrastructure or agent platforms used by multiple customers.Experience with distributed systems and event-driven architectures.Experience implementing robust agent evaluation frameworks.Contributions to open-source AI/agent frameworks or infrastructure.Strong understanding of LLM limitations, failure modes, and emerging agent architectures.Experience taking an AI system from prototype → production → scale.IMPORTANT:This is not a prompt-engineering role.We are looking for someone who can architect and engineer reliable AI systems end-to-end.If your experience is primarily building ChatGPT wrappers, writing prompts, or experimenting with AI tools without deploying production-grade agent systems, this role is unlikely to be a fit.What You'll Get To Build;Production-grade AI agent systems from the ground up.Complex single-agent and multi-agent architectures.Autonomous workflows connected to real-world systems and APIs.AI infrastructure designed for scale, reliability, and security.Advanced RAG, memory, evaluation, and orchestration systems.Systems that combine LLMs with asynchronous infrastructure, automation, and external tools.This is a fully remote opportunity and applications are open worldwide.