Sergei Belousov

I’ve been obsessed with computer vision, machine learning, and all things AI for over a decade — and I still genuinely love it.

[Sergei Belousov @bes_dev]

@rra @NASA Working on RS-25 test software at MSFC — legend status! To be honest, building a 5-year-old React project today is harder than compiling this 56-year-old FORTRAN IV 😄 My favorite find: when altitude goes negative, SSSP prints VEHICLE ATTEMPTED SUBTERRANEAN FLIGHT. Honored!

[Sergei Belousov @bes_dev]

How was your weekend? I spent mine launching the Space Shuttle. With NASA code from 1970 that nobody could run for 50 years.

[Sergei Belousov @bes_dev]

@planetdeimos That is incredibly cool. It’s a huge privilege to know that the engineers working on SLS might find this weekend project interesting. This code was decades ahead of its time, and I’m just glad to help keep that history alive. Thanks for sharing it, Sean!

[Sergei Belousov @bes_dev]

@reach_espartan @NASA More detailed information and documentation can be found in the GitHub repository: https://t.co/hMZz4aE7zz. I also provided a general description of the math behind the simulator, based on my analysis of the code.

[Sergei Belousov @bes_dev]

@aerodud3 @DJSnM Of course, we're not talking about Fortran programs in general, but about a specific pre-project Shuttle simulator.

[Sergei Belousov @bes_dev]

@reach_espartan @NASA They used the fifth-order Runge-Kutta-Butcher method with adaptive stepping.

[Sergei Belousov @bes_dev]

@MKortvely Wow, thanks for sharing this story! Running a simulation and knowing a shuttle is flying above you along that exact calculated trajectory is practically magic. Especially when you consider the era and the limitations engineers faced building this software.

[Sergei Belousov @bes_dev]

@samuel_hayden_ @macjshiggins To clarify: the "test data" is actual STS-1 telemetry from 1981 vs a theoretical 1970 model. Getting only a 5-15% delta between 1970 math and a physical flight 11 years later just proves the absolute brilliance of the engineers of that era given their limited resources.

[Sergei Belousov @bes_dev]

@macjshiggins I've published all the necessary code, instructions, and experiment logs to reproduce this simulation in the public domain. If you encounter any issues reproducing it, please let me know.

https://t.co/hMZz4aE7zz

[Sergei Belousov @bes_dev]

7,811 lines. 44 subroutines. 10 fixes. Wikipedia. One `make run`.

https://t.co/wE0ZcIFcQN
https://t.co/hMZz4aE7zz

@NASA

[Sergei Belousov @bes_dev]

The RS-25 engines in this simulation are the same engines flying RIGHT NOW on SLS for Artemis. The code that evaluated whether the Shuttle concept would work - still gets the math right for hardware heading to the Moon.

[Sergei Belousov @bes_dev]

Spent a weekend vibe-coding a massive legacy codebase with Claude Code. The trick: a supervisor agent that builds a harness before each task and learns from session logs after. Blogpost: https://t.co/B4owUjsdRi

#ai #aiagents

[Sergei Belousov @bes_dev]

@om_patel5 There is similar project (AI agent for full stack mobile app development, with extended QA capabilities including UI tests on emulator and real devices): https://t.co/I776xXPsyJ
I wrote blogpost about how my QA Engoneer works: https://t.co/P58RmZSj43

[Sergei Belousov @bes_dev]

AI can write apps and tests. But who actually opens the app and taps the buttons?

I built a QA agent that tests UI promises, not code. It taps through the app on a simulator, reads screenshots, and catches bugs unit tests can't see.

Blogpost: https://t.co/P58RmZSQTB

[Sergei Belousov @bes_dev]

Open-sourcing my local alternative to @rork. Been using it to ship production apps.

Etnamute — AI mobile developer on top of @claude_code . Idea → interview → market research → build with QA → App Store deploy.

No framework, no extra subscription.

https://t.co/Qf8y2WO8oP

[Sergei Belousov @bes_dev]

Finally scratched an old itch — AI sticker packs from your photos. Works with Telegram/WhatsApp. 👉 https://t.co/9k6QTWR3l0