Senior DevOps Engineer
Posted 2
views
Summary
Senior DevOps engineer on Nayatel's AI Platform team, building and operating Azure cloud infrastructure, an LLM gateway, and model serving behind a wealth management product. Day-to-day is hands-on infrastructure-as-code (Terraform/Bicep), cloud networking, reliability, and cost optimization.
Job Overview:
You’ll join the AI Platform team building and operating the cloud infrastructure, LLM gateway, and serving layer behind our wealth management product. This is a hands-on ownership role on Azure for an engineer who can take infrastructure from code to production and keep it running reliably and cost-effectively.
Requirements:
- 3-5 years of platform / backend / infrastructure engineering, with real ownership of things running in production.
- Hands-on cloud experience with any major provider - Azure, AWS, GCP, or equivalent -including a solid grasp of core compute and networking (private networking, container/serverless compute, and how traffic flows between services).
- Infrastructure as code with Terraform, Bicep, or a comparable tool - modules, state, and the discipline of managing infra in code.
- Solid debugging instincts - you reach for logs, traces, and reproduction steps before guessing.
- Working knowledge of LLMs: what tokens, context windows, and quotas mean in practice, and how those constraints shape what you can build.
- Fluency with modern AI-assisted development tools - Claude Code, Cursor, Antigravity, OpenCode, or similar.
Nice to Have:
- Hands-on Azure specifically - VNets, subnets, private endpoints, Container Apps, Function Apps.
- System design experience, familiarity with AI infrastructure terms (tokenization, requests per minute, token budgeting)
- Strong Python, including async (asyncio), and a working grasp of concurrency, threading vs.async, and the pitfalls of each.
- Experience building or operating an LLM gateway or similar proxy/routing layer.
- Open-source model deployment and serving - running and hosting your own models (vLLM, Ollama, TGI, or similar), and the infra/optimization that comes with it.
- Exposure to LLM safety and governance - PII redaction, data classification, prompt/response guardrails.
- Cost optimization across cloud compute (plans, scaling, cold starts on serverless).
- Observability tooling (App Insights, OpenTelemetry, or similar)
