AI Platform SRE for Banking-Grade Resilience

Deeplight; Abu Dhabi; On-site; AED 260,000 - 460,000; Full time

Benefits offered by this job

Visa sponsorship

Comprehensive health insurance

Professional development support

Monthly incentive program

Career advancement opportunities

Cutting-edge AI projects

Job summary

DeepLight AI seeks a Site Reliability Engineer to own production readiness, resilience, and operational health of AI solutions in the UAE. You will design load testing, chaos engineering, and automated testing for LLM and RAG models, ensuring scalable, secure, banking-grade performance under enterprise traffic.

You will build CI/CD regression tests, implement observability dashboards, and set SLIs/SLOs with real-time telemetry.

Qualifications

• Proven experience building and testing AI/ML platforms at scale.
• Strong Python and Bash scripting ability.
• Experience with cloud providers (Azure/AWS) and IaC.
• Familiarity with CI/CD and automated testing.
• Knowledge of RAG, LLM evaluation metrics, and model quality.

Responsibilities

• Design load tests and chaos engineering to sustain banking-scale traffic.
• Develop Python-based regression tests integrated into CI/CD.
• Define SLIs/SLOs and implement real-time telemetry and alerts.
• Integrate AI evaluation harnesses into deployment pipelines.
• Enforce security scans and data residency compliance.

Skills

Python

Bash scripting

Automation testing

CI/CD

IaC (Terraform)

Cloud (Azure/AWS)

Kubernetes (AKS/EKS)

Monitoring/Observability

MLOps / RAG evaluation

Security/compliance

Tools

GitHub Actions

Azure DevOps

Terraform

K6 / JMeter

Prometheus / Grafana

Job description

DeepLight AI seeks a Site Reliability Engineer to own production readiness, resilience, and operational health of AI solutions in the UAE. You will design load testing, chaos engineering, and automated testing for LLM and RAG models, ensuring scalable, secure, banking-grade performance under enterprise traffic.

You will build CI/CD regression tests, implement observability dashboards, and set SLIs/SLOs with real-time telemetry.