Senior Network Software Engineer – Python & Go

SHIMI Sp. z o.o.

• Senior expertise in Python and Go focusing on network software.
• Deep understanding of networking layers L2-L4 and protocols including IPFIX, NetFlow, sFlow.
• Experience with async programming internals in Python (asyncio).
• Proficiency in static typing and code quality tools (mypy, golangci-lint).
• Strong Linux system internals and performance profiling skills.
• Knowledge in containerization and Kubernetes.
• Familiarity with telemetry data streaming using Kafka and timeseries databases like ClickHouse.
• Ability to automate network device configuration across multiple vendors.
• Experience with Protocol Buffers schema management.
• Familiarity with IPC and distributed state design patterns.

Nice to have:

• Experience with eBPF.
• Knowledge of Windows Active Directory integration.
• Experience with cloud SDKs for Azure or AWS.

Core languages:

• Develop concurrent services and UDP/TCP servers in Go with profiling.
• Implement async programming with Python using asyncio, including HTTP client/server.

Quality discipline:

• Use static typing tools like mypy and golangci-lint, formatters, linters, and pre-commit gates.
• Write async unit tests, snapshot tests, and load tests integrated in CI.
• Maintain dependency hygiene and reproducible pinned builds.

Networking depth:

• Handle L2/L3/L4 headers manually, VLANs, MAC/IP semantics, TCP flags, MTU/fragmentation, IPv4/IPv6.
• Work with flow telemetry protocols: IPFIX, NetFlow v5/v9, sFlow v5 including template decoding and loss handling.
• Capture and craft packets; use pcap for debugging.
• Automate network devices over SSH/CLI and vendor APIs (Cisco, Arista, Juniper, Palo Alto) with config drift detection.

Data movement:

• Design Protocol Buffers schemas and codegen; implement fast JSON/binary codecs.
• Handle Kafka producer/consumer semantics including partitioning, compression, SASL/Kerberos.
• Work with columnar/timeseries sinks like ClickHouse.

Performance engineering:

• Perform routine profiling focused on allocation, GC pressure, lock contention, zero-copy paths.
• Implement data structures such as LPM/Patricia tries, LRU caches, and hashing with tradeoff explanations.
• Conduct load generation and throughput testing.

Systems and delivery:

• Utilize Linux internals (sockets, epoll, sysctl tuning, cgroups).
• Work with eBPF (strong plus).
• Develop using containers and Kubernetes with multi-stage builds.
• Deliver self-contained binaries with packaging and upgrade/rollback support.
• Use IPC patterns like ZeroMQ.

Distributed state:

• Design cache/state synchronization with eventual consistency and reconnect behavior.

Identity and cloud integrations (plus):

• Integrate with Windows/AD technologies (WinRM, SMB, Kerberos).
• Use cloud SDKs (Azure or AWS) for asset and log ingestion.