Senior Backend Engineer (AI Agents)

GCS; Abu Dhabi; On-site; AED 300,000 - 600,000; Full time

Benefits offered by this job

Competitive rates

Onboarding in coming weeks

Job summary

GCS is seeking a backend engineer to architect and implement the orchestration layer that powers AI agents in production. You’ll design scalable, containerised microservices with strict security, RBAC and PII safeguards, connecting LLMs to ERP data in on-prem environments.

You will work on durable, auditable systems at high concurrency, integrating with SAP S/4HANA/Ariba and deploying across on-site and air-gapped setups.

Qualifications

• 5+ years in backend engineering for distributed AI/ML in production.
• Expert Python (FastAPI or Django) and Go/Java for high-performance services.
• Strong Docker and Kubernetes experience, ideally on‑premise.

Responsibilities

• Architect the orchestration layer for AI agents and task routing.
• Build scalable microservices with secure data access to ERP systems.
• Design secure gateways and RBAC enforcing PII redaction before LLM access.

Skills

Python

Go/Java

Docker

Kubernetes

REST/gRPC

API Gateways

RBAC/Security

Distributed Systems

Tools

Kong

NGINX

SAP S/4HANA

Ariba

Job description

About the opportunity

We’re partnered with a leading sovereign AI company in the UAE, delivering a flagship AI transformation within defense.

This is a rare backend role at the heart of enterprise AI. You’ll build the “nervous system” that powers AI agents in production: the orchestration layers, secure API gateways and integrations that connect Large Language Models (LLMs) to industrial ERPs and sensitive engineering data. Production systems, not demos.

You’re not just building APIs; you’re building the “hands” that let AI do real work.

What you’ll do

• Architect the “Agent of Agents” orchestration layer that coordinates specialised AI agents, managing state, task routing and error handling
• Build the task router and intent classifier services that map requests to skills and playbooks, with deterministic execution for critical financial and engineering workflows
• Design scalable, containerised microservices for high-concurrency, low-latency workloads
• Build the secure Data Access Layer that lets AI agents read from and write to SAP S/4HANA and Ariba, with strict validation before any transaction is committed
• Connect agents to vector databases, retrieval models and external data feeds, handling caching, rate limiting and normalisation
• Implement gateway and policy services that enforce authentication, RBAC and PII redaction before data ever reaches an LLM
• Engineer for on-premise and air-gapped deployment, with full logging, tracing and observability

What we’re looking for

• 5+ years in backend engineering, building distributed systems or platforms that serve AI/ML in production
• Expert Python (FastAPI or Django) and Go or Java for high-performance services
• Strong Docker and Kubernetes experience, ideally in on-premise environments
• RESTful and gRPC API design; experience with API gateways (Kong, NGINX)
• Experience building high-availability, auditable systems, ideally in fintech, healthcare or defence

Background we love (not required, but a strong signal)

Hands-on work with AI agents, LLM orchestration or MCP; ERP integration (SAP, OData, SOAP); or delivering in regulated, structured environments.

How the team works

• An elite team of engineers from FAANG and frontier AI labs
• What matters is what you ship, not the hours you spend
• Low ego, high standards. The best idea wins, whoever it comes from
• Rigorous engineering, delivered at startup pace

The setup

• Competitive rates
• Onboarding in the coming weeks

Why this role?

Your code will enable AI agents to forecast critical supply shortages, validate the safety of engineering designs, and automate high-stakes financial workflows.

You’ll be one of the few engineers in the world defining how Large Language Models connect to the industrial core of a defence prime, turning static systems of record into dynamic systems of intelligence. If you want to build the secure, high-performance architecture that makes AI tangible, this is the role.