Senior AI Agent Developer (LLM Systems & Workflow Automation)

Location

Nigeria

Work Model

Remote · Autonomous Execution · Contract or Full-Time (project-based start)

About the Role

We are building a production-grade AI customer support system for a large medical e-commerce platform operating on Shopify Plus.

This role goes far beyond a traditional chatbot. You will design and implement an AI agent system that combines retrieval-augmented generation (RAG), workflow orchestration, and multi-LLM routing to automate real customer support operations end-to-end.

The chatbot, workflows, and integrations function as one unified system, handling product inquiries, order tracking, ticket management, and intelligent escalation.

What You’ll Be Working On

• AI chatbot with RAG-based knowledge retrieval
• Workflow automation using n8n (or equivalent orchestration tools)
• Multi-LLM orchestration for cost, latency, and task optimization
• Shopify Plus API integrations (products, orders, customers – read-only)
• Gorgias helpdesk integration (ticket creation and updates)
• Production deployment and monitoring on Google Cloud

Key Responsibilities

• Build AI agents that call tools, APIs, and workflows, not just generate text
• Implement robust RAG pipelines across product data, FAQs, policies, and custom documents
• Design and deploy n8n workflows for intent routing, order lookups, and ticket handling
• Implement multi-model LLM strategies based on task complexity and cost
• Own deployment, monitoring, logging, and token optimization
• Ensure system reliability with strong error handling and fallback logic

Required Technical SkillsAI Agent Development (Critical)

• Proven experience building production AI agent systems
• Function calling / tool use with LLMs
• ReAct or similar agent architectures
• Prompt engineering for consistent, deterministic outputs
• Handling malformed or unexpected LLM responses

Evidence expected: GitHub repos or real-world systems demonstrating LLM-driven workflows.

Retrieval-Augmented Generation (Critical)

• Embedding model selection and optimization
• Document chunking strategies
• Retrieval tuning (similarity thresholds, reranking, metadata filtering)
• Experience with vector databases (Qdrant preferred)

Knowledge sources include product catalogs, FAQs, policy documents, and custom context files.

Workflow Orchestration (Important)

• Experience with n8n (preferred) or similar tools (Temporal, Airflow, etc.)
• REST APIs, webhooks, and automation patterns
• Error handling, retries, and fallback strategies

Multi-LLM Strategy (Important)

• Experience with OpenAI APIs (GPT-4, GPT-3.5)
• Experience with Anthropic Claude
• Familiarity with open-source models (Llama, Mistral, etc.)
• Understanding cost, latency, and capability trade-offs

Production Engineering (Important)

• Self-hosted deployments
• Google Cloud Platform familiarity
• Monitoring, logging, and performance optimization
• Token usage tracking and cost control

Working Style

• High autonomy and ownership
• Async-first communication via Slack
• Weekly sync calls
• Collaboration with AI/Product and Lead Engineering stakeholders
• Europe-friendly working hours preferred

Who This Role Is For

This role is ideal for someone who:

• Has already shipped AI agent systems into production
• Understands LLMs as controllers of workflows, not just chat interfaces
• Is comfortable owning architecture, implementation, and reliability

This role is not suitable for junior developers or prompt-only chatbot builders.

How to Apply

CVs should be sent to:

📧 HRNG@TURBHAM.COM

Please include links to relevant projects or repositories demonstrating AI agent or LLM workflow experience.

<