The role

You are our Azure authority. Everything that touches Azure architecture, build, operations, security, cost, and compliance stops with you. We need to be able to hand you an Azure question, any Azure question, and trust the answer without a second opinion.

Concretely: we run a multi-agent AI system for a large enterprise client. Our engineering team designs the agents and back-end services and knows their way around Azure but youre the expert, and on Azure your call is the one that stands.

You set the implementation direction: which services, what topology, how it’s secured, how it scales, what it costs. Then you stand it up, harden it, and run it in production. Today it’s one agent. The target is a fleet with real SLAs, spanning Azure-hosted services and the Copilot surface.

You’re also the only technical voice in two rooms you can’t delegate. On one side, a client business team with no technical background who need to understand what you’re building and why. On the other, the client’s security and governance organisation, whose compliance findings and remediation deadlines land on your desk and nobody else’s.

This is a broad role by design. If you’re deep in one Azure service and thin everywhere else, this isn’t it.

Platform you’ll design and operate

•Compute & hosting: Container Apps or AKS, App Service, VMs and scale sets, serverless where it fits — you pick per workload and justify it

•AI layer: Azure AI Foundry, model deployments, quota and PTU planning, content safety, evaluation pipelines, Azure AI Search for retrieval, embedding and index lifecycle

•Data: Cosmos DB,



Azure SQL or PostgreSQL, Blob/Data Lake, Redis — plus agent state, memory, and conversation stores

•Orchestration & messaging: Service Bus, Event Grid, Event Hubs, durable workflows, retry and dead-letter design across agents

•Networking: VNets, private endpoints, Private DNS, NSGs, App Gateway/WAF, Front Door, egress control, hybrid connectivity

•Identity & secrets: Entra ID app registrations, managed identities, RBAC, Key Vault, certificate and secret rotation

•Observability: Log Analytics, App Insights, distributed tracing across agent hops, token and cost telemetry, alerting, SLOs

•Platform engineering: Bicep or Terraform, landing zone and subscription design, CI/CD, setting strategy, DR and backup

•Copilot surface: Copilot Studio and M365 Copilot agents, Power Platform environments and DLP policy, Dataverse, connectors — and orchestration between those and the Azure-hosted agents

•Cost: capacity planning and FinOps for a growing agent fleet, not a fixed workload

Working with our engineering team

•Take the architecture our engineers design and decide how it gets built on Azure: which services, how they connect, what it costs, where it breaks





•Push back when a proposed design doesn’t map cleanly to Azure, and propose the alternative

•Build the reference implementation and deployment patterns the team codes against, so they can ship without you in every decision

•Set the guardrails: environment structure, secrets handling, networking, identity, what engineers can provision themselves

•Review their work for performance, cost, security, and compliance before it hits production

Microsoft governance — hard requirement

You’ll absorb the compliance and security pressure: Azure Policy non-compliance notices, Defender for Cloud findings, SFI (Secure Future Initiative) requirements, S360-style remediation deadlines, subscription and tagging governance, admin consent and app approval processes.

We need someone who has worked inside Microsoft’s own tenant and subscription environment — as an employee, vendor, or supplier — and knows both the technical fix and the process to get it approved. Generic enterprise Azure experience does not cover this.

Communication — equally hard

•Explain architecture, trade-offs, and risk to stakeholders with zero technical background, in plain language

•Translate a security notice into “here’s the impact, here’s what I need from you, here’s the date”

•Find the right owners inside a large organisation and get unblocked without escalating everything back to us

•Produce architecture artefacts — diagrams, decision records, runbooks — that engineers and business stakeholders can both use

📌 Senior Azure Platform Engineer - AI Agent Infrastructure (Bengaluru)
🏢 Truvs
📍 Bengaluru