about me

Rishat Maksudov

Software Engineer (C/C++ Developer) - Tashkent, Uzbekistan

Systems programmer with 4+ years of experience building high-performance network and storage software in C/C++. At Huawei worked deep in the Linux networking stack using DPDK for packet processing. Currently at Wellstor working on storage systems with SPDK. Care about clean low-level code, performance, and understanding how things actually work.

experience

July 2021 - November 2024

Junior Software Engineer

Huawei

Worked on router IoT/SOHO/AP firmware (CLI, NAT subsystem) for the first half, then moved to eNSP Pro (Lite) - Huawei's enterprise network simulation platform - building virtual wireless infrastructure with eBPF and NFQ.

  • Maintained and extended legacy NAT subsystem in C - refactored NAT rule handling and fixed ALG (Application Layer Gateway) parsing for PPTP and H.323 protocols where standard port-based NAT was insufficient
  • Fixed NAT ALG packet classification bugs where L4 port parsing was bypassed in favour of application-layer inspection - restored correct traversal for PPTP, FTP, and DNS protocol flows
  • Built virtual wireless network infrastructure for eNSP Pro (supports ~20 concurrent virtual devices) using NFQ (nfqd) and eBPF - software-defined packet interception and forwarding between virtual interfaces
  • Implemented roaming logic for virtual mobile clients between access points - proximity-based AP selection with seamless handoff across the virtual topology
  • Designed and implemented ebridge - a virtual Ethernet bridge connecting interfaces across two devices into a shared L2 segment, supporting arbitrary interface counts

January 2025 - present

Software Engineer

Wellstor

Building block storage software on SPDK - RPC disk management, inter-node heartbeat for configuration consistency, and NVMe/io_uring device handling.

  • Fixed critical blocking bug in disk heartbeat - TCP NVMe-oF connect was synchronous, causing up to ~2 minute stalls per failed node (6 retries × 20s timeout); replaced with non-blocking connect and immediate failover
  • Implemented RPC command layer for NVMe and io_uring disk management via SPDK - device discovery, hot-plug handling, and lifecycle control
  • Designed inter-node disk heartbeat for configuration synchronisation - coordinates disk topology, splits, and RAID state across cluster nodes before backup commits
  • Built disk health validation tooling - sequential and random read/write tests with per-disk pass/fail reporting

education

September 2014 - May 2017

Secondary education

Academic Lyceum under Westminster University in Tashkent

Exact sciences

September 2017 - May 2021

Bachelor's degree

Innopolis University

Software Engineering