Skip to content
View mohammmad357's full-sized avatar
😃
Happy face and broken inside
😃
Happy face and broken inside
  • Iran
  • 14:31 (UTC +03:30)

Block or report mohammmad357

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

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

Report abuse
mohammmad357/README.md

Experiments you can inspect. Results you can reproduce.

Python · C11 · Algorithms · Storage recovery · Simulation

I build small open-source engineering projects with readable implementations, independent checks and documented boundaries.

Pathfinding · Storage · Simulation · Inverter control

Start here

Project The question Evidence to explore
Maze Signal When is a longer route cheaper, and when does A* explore less? Python/C parity, 500 seeded maps checked against Bellman-Ford, reproducible search footprints
CrashVault What survives after an acknowledged write and a process restart? Matching binary logs, truncation/corruption checks, explicit sync and recovery contracts
Faultline Can two languages reproduce the same stochastic event history? 300 paired trajectories, independent replay, analytical checks and raw experiments
PowerControlLab How do inverter controllers behave under defined disturbances? Simulation models, controller derivations, benchmark protocol and an experiment dashboard

Try Maze Signal first: follow its quick start, then inspect the measured results.

How I approach a project

  • Define a small problem and its input contract.
  • Check behavior against an independent reference or mathematical invariant.
  • Publish commands, raw measurements and the conditions behind the numbers.
  • Explain limitations alongside results; these projects do not establish production readiness.

The implementations were developed with AI coding assistance. Claims are tied to executable tests and documented experiments.

Current direction

Try the published Maze Signal demo: draw weighted maps and compare Dijkstra with A*. The owner confirmed the trap costs 8 in both panels; a later in-app-browser check encountered a DNS failure. If Pages is unavailable on your network, use the repository quick start.

Visual example: four moves cost 28, eight cost 8 · Goldbach bug story · Portfolio evidence and interview cards · Interim progress report · How to support the work

Two proposed upstream fixes are open as drafts, not merged contributions: parallel sorting #15554 and repeated-prime Goldbach pairs #15555. Current priority: real first-use feedback and small maintenance work.

If you try a project, open an issue in that repository with the command, environment, expected behavior and actual result. Small, well-explained contributions are welcome.

فارسی — دربارهٔ این مجموعه

این مجموعه برای یادگیری و نمایش مهندسی قابل‌بررسی است: الگوریتم، بازیابی داده و شبیه‌سازی در Python و C. برای شروع، Maze Signal را اجرا کن و بعد نتیجه‌ها و آزمون‌ها را ببین. تمرکز فعلی روی استفادهٔ آسان‌تر، بازخورد واقعی و نگهداری پروژه‌هاست. CrashVault یک موتور آموزشی کوچک است؛ Faultline پیش‌بینی بیماری نیست؛ نتایج PowerControlLab فعلاً شبیه‌سازی‌اند.

Support / حمایت

Support maintenance on Coffeebede — Iranian domestic payments

Voluntary support helps maintain these educational projects and their documentation. This is a domestic payment option; international card and cryptocurrency payments are not available here. No paid support tier or delivery deadline is promised.

حمایت ریالی از نگهداری پروژه‌ها — این لینک درگاه داخلی است و پرداخت با کارت خارجی یا رمزارز ندارد.

QR: Coffeebede domestic support

Scan for the domestic payment page / اسکن برای صفحهٔ حمایت ریالی.

Pinned Loading

  1. crashvault crashvault Public

    An acknowledgment is a promise. Interoperable Python & C11 storage engines with a checksummed append log, guarded framing, restart recovery, fault-injection tests and reproducible benchmarks.

    Python

  2. faultline faultline Public

    One network. Two engines. The same stochastic history. Reproducible Gillespie SIR simulation in Python and C, verified event by event, with measured experiments and sanitizer CI.

    Python

  3. maze-signal maze-signal Public

    The shortest route isn't always the cheapest. Optimal weighted pathfinding in Python & C, with A*, Dijkstra, visual comparisons, differential tests and reproducible benchmarks.

    Python 1

  4. PowerControlLab PowerControlLab Public

    Reproducible DC-AC inverter control benchmarks in Python: PI, SMC, adaptive SMC and constrained MPC, with SPWM, nonlinear loads and an interactive dashboard.

    Jupyter Notebook