Skip to content
View FI-Mihej's full-sized avatar

Highlights

  • Pro

Block or report FI-Mihej

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
FI-Mihej/README.md

FI-Mihej

Python & C/C++ Team Lead
Building high-performance OSS systems
Async • multiprocessing • shared memory • LLM infrastructure


Focus

  • High-performance Python systems (beyond typical GIL-bound design)
  • Inter-process communication (IPC) with minimal overhead
  • Shared memory architectures
  • Async runtimes and event loop engineering
  • Systems-level optimization (C/C++ / Cython / low-level integrations)

Open Source Contributions

Contributor to:

Authored a fix for an important Windows issue in Kivy's core (included via an upstream PR).


Key Projects

High-performance IPC via shared memory.

  • Zero/low serialization overhead approach
  • Seamless sharing and mutation of Python objects across processes
  • Supports:
    • NumPy arrays
    • Torch tensors
    • custom classes (including dataclasses)
    • objects with methods
    • asyncio integration

Core idea:
Move away from serialization-heavy IPC → toward shared-memory object models with controlled mutation.


Multi-domain high-performance Python toolkit.

Includes:

  • Runtime bytecode manipulation
  • True shared memory primitives
  • Async integrations:
    • LMDB
    • Tkinter
    • wxPython
    • PySide / PyQt
  • Custom async loop with near preemptive multitasking (single thread)
  • Advanced introspection & text parsing tools

Implementation stack:
Python + Cython + C/C++ + Nim + Go (where performance matters)


Engineering Direction

  • Reducing Python overhead via architecture, not micro-optimizations
  • Treating IPC as a first-class performance domain
  • Bridging high-level ergonomics with low-level control
  • Designing systems that scale beyond "standard async patterns"

Tech Stack

Languages:
Python, C/C++, Nim

Domains:

  • Async systems
  • IPC / shared memory
  • Systems programming
  • Performance engineering

Contact / Work

  • Open for high-impact system-level work
  • Focus: performance-critical backends, infra, unconventional architectures

Pinned Loading

  1. zauberzeug/nicegui zauberzeug/nicegui Public

    Create web-based user interfaces with Python. The nice way.

    Python 15.6k 910

  2. InterProcessPyObjects InterProcessPyObjects Public

    High-performance and seamless sharing and modification of Python objects between processes, without the periodic overhead of serialization and deserialization. Provides fast inter-process communica…

    Python 88 3

  3. Cengal Cengal Public

    Bytecode manipulation in runtime, true shared memory, async LMDB, async Tkinter, async wxPython, async PySide, async PyQt, async loop with almost preemptive multitasking within the single thread, t…

    Python 31 3

  4. Bensbach Bensbach Public

    Unreal Engine 3 (UE3) compiler and decompiler (for games modding). Convenient, simple language for those people who knows what UE3 bytecode is. And easily understandable for others.

    Python 14 4

  5. cengal_app_dir_path_finder cengal_app_dir_path_finder Public

    A Python module offering a unified API for easy retrieval of OS-specific application directories, enhancing data management across Windows, Linux, and macOS

    Python 3

  6. FeatherCoroutine FeatherCoroutine Public

    A lightweight C library for memory-efficient asynchronous programming with coroutines and a microkernel-inspired main loop, ideal for constrained environments such as iOS background tasks

    C 1