Azure Architect

Job Description

Architecture & Reliability Expectations (Event-Driven Systems)

Ensure Real-Time Event Pipeline Stability

Design and enforce idempotent event processing, at-least-once or exactly-once delivery semantics

Currently Duplicate detection is enabled at Azure Service Bus Infrastructure level, and at Application-level messages being validated against duplicate processing.

Manage dead-letter queues (DLQ) and retry policies

Define event ordering, deduplication, schema evolution

Maintain SLOs for latency, throughput, and event loss. Define the same for new designs coming up for upcoming Transaction Groups.

Event backlog monitoring

Consumer lag thresholds

Circuit breaker & back pressure strategies

Event loss = 0

Reprocessing success rate > 99.9%

End-to-end event latency within SLA

Since the environment is not yet fully Active-Active, the architect is expected to own the phased maturity roadmap.

Front Door/APIM routing

Stateless compute enablement

Geo-replicated messaging (Azure Service Bus)

Cosmos DB multi-region writes

Azure SQL Hyperscale read replicas / failover groups

Events Rejected & Reprocessing Governance

Expectation:
Design cross-region failover strategy (Example: EA2 → Central US)

DB2 connection resiliency validated through DR drills

Expectation:
Architect geo-redundant Function Apps

Cross-region deployment pipelines

Storage replication (GRS / RA-GRS)

Mainframe failover connectivity strategy

Document file ingestion failover workflow

Functions can fail over to Central US with minimal downtime

File ingestion resumes without data loss

Event replay pipelines

Audit trail for rejected events

Data consistency & reconciliation logic

Replay throttling & idempotency safeguards