Senior Python Engineer – Network Orchestration (DriveNets)

About DriveNets

DriveNets is a rapidly growing software company that has created a radical new way for service providers and hyperscalers to build their networking infrastructure. DriveNets Network Cloud and DriveNets Network Cloud-AI are new innovative networking solutions that apply the cloud architectural approach to high-scale networking. They bring together the scalability of standard Ethernet Clos architecture with the high performance and reliability of service provider networking, delivering optimal networking performance, scale and cost structure for service providers and hyperscalers.

The Role

We're building a cloud-native network orchestration platform running on Kubernetes, responsible for automating the full lifecycle of devices and services across tier-1 ISP router networks — provisioning, modification, reconciliation, and teardown of BGP, IS-IS, and MPLS L2/L3 VPN services at carrier scale.
You'll join the workflow engineering team as a network-orchestration engineer with strong Python skills. Your primary job is to take a provisioning blueprint or service-design specification from our System Architects and implement it.
The role spans a service lifecycle end-to-end — spec, planner, workflow, rollback, tests.
This is a backend, systems-oriented role focused on correctness, reliability, and operational robustness at carrier scale.

Zakres obowiązków

• Read provisioning blueprints and service-design specifications; translate them into Python plan generators that produce device configuration and service-related artifacts
• Pair with System Architects on unclear spec areas; drive clarifications back into the spec and the code together
• Model service-lifecycle behaviour: IP-Link / BGP / IS-IS / sys-config service graphs, with create-vs-adopt semantics, idempotency, and undo
• Design Temporal-based workflows and custom Python tasks that consume your plans and drive the network safely, handling partial failures without leaving the system in an unrecoverable state
• Contribute to reference-data alignment, configuration-template generation, and workflow YAML authoring as your feature demands
• Integrate with platform systems (inventory, IPAM, resource manager, service catalog) and with router management interfaces (NETCONF, gNMI, REST, CLI) via the platform's abstractions
• Land your changes across multiple repositories and long-lived release branches, including porting fixes between them
• Add observability — metrics, structured logs, workflow-instance traces — that makes a stuck workflow diagnosable from Temporal alone
• Write and maintain tests: pytest, golden fixtures, tycheck-clean code; contribute to E2E scenarios where relevant
• Use our AI-native development pipeline (spec → design → plan → code → review → close) as your daily working model

Wymagania

Required qualifications

• Service-provider networking: hands-on with BGP, IS-IS, MPLS L2/L3 VPNs, IP routing, addressing/IPAM concepts. You hold a design discussion about a service model without a translator
• Strong Python: production-quality code — clear APIs, structured error handling, tests. Python is your working tool. 3+ years in a production codebase
• Spec-to-code: you read a design spec, turn it into code an architect recognises as correct, and push back when the spec needs iterating
• Working proficiency with Kubernetes and cloud environments (Azure or AWS): you deploy, run, and troubleshoot services in a K8s cluster
• Troubleshooting: you correlate events and logs across several distributed systems to find root cause
• SQL databases as an application developer — migrations, schemas, transactions
• Fluent English — the team is international

Nice-to-have qualifications

• Hands-on with a workflow-orchestration engine
• Experience owning a feature end-to-end including its rollback / recovery path — you've designed "safe to fail" and not only "safe to succeed"
• Familiarity with router platforms in a service-provider context (Cisco, DriveNets/DNOS, Juniper, Nokia, Arista)
• NETCONF/YANG or gNMI/OpenConfig exposure
• Observability stack experience — you've instrumented services in Prometheus/Grafana or equivalent
• Comfort in Linux, Docker, and virtualised environments as a daily driver
• Experience with CI/CD pipelines and multi-repo release engineering
• Experience operating an AI-assisted development pipeline: agent-driven code review, LLM-drafted specs, orchestrated ticket flows

Tech stack

• Python
• Data base
• K8s
• Cloud
• Networks