I am a systems architect with 10+ years in systems integration, middleware, distributed platforms, and production operations. I work across the full operating path: topology, identity, deployment, observability, failure handling, and recovery.
I treat failure modes, operator actions, and client-visible verification as architecture inputs rather than post-deployment tasks.
| Area | Work |
|---|---|
| Distributed streaming | Kafka/KRaft topology, quorum design, replication, min.insync.replicas, capacity, upgrades, and failure boundaries |
| Integration and middleware | Event-driven integration, CDC, connectors, APIs, message contracts, IBM MQ, and IBM Integration Bus |
| Identity and policy | TLS/mTLS, SASL, Kerberos, OAuth/OIDC, ACL/RBAC, secret boundaries, and fail-closed automation |
| Production operations | Observability, capacity planning, runbooks, rolling changes, backup/DR decisions, restore tests, and incident recovery |
Apache Kafka · KRaft · IBM MQ · IBM Integration Bus · Go · Python · Linux · PostgreSQL · Docker · Kubernetes · Ansible · Prometheus · Grafana
- Start from failure domains, authority boundaries, and recovery objectives.
- Bind documentation claims to runnable checks or explicit evidence gaps.
- Treat configuration, apply, readback, and rollback as one change contract.
- Verify the real client path after infrastructure health returns.
- Rehearse failure and clean-host recovery before relying on a design.
My current work focuses on governed AI runtimes and operator controls: isolated execution, explicit permissions, multi-provider routing, checkpoints, evidence, and recovery.
Generated from public GitHub events. Private repository activity is excluded.
For systems architecture, middleware, platform engineering, or technical advisory work: LinkedIn · kavioavio.ru