Dominik Filkus

⚙️ Tech lover | 🖥️ Software Engineer | 🕹️ Retro games to AI innovation

[Dominik Filkus @DominikFilkus]

I made this Super Pang game in just 3 prompts with GPT-5.4. This model is really good at generating games.

[Dominik Filkus @DominikFilkus]

@alojohhardcore Sold Amazon for this one

[Dominik Filkus @DominikFilkus]

@EricBalchunas Corruption has stopped 🎉

[Dominik Filkus @DominikFilkus]

@alojohhardcore If I hold the stock for another 5 years I might get a lollipop.

[Dominik Filkus @DominikFilkus]

@AlexIvyson @alojohhardcore @petemesquete @LarenLaurens This is the point I think.

[Dominik Filkus @DominikFilkus]

@alojohhardcore Simple and professional. This is why I like HC so much.

[Dominik Filkus @DominikFilkus]

@ClaudeDevs So this is why workflow magic word has stopped working

[Dominik Filkus @DominikFilkus]

@magikaarpe @alojohhardcore Satire ofc :)

[Dominik Filkus @DominikFilkus]

@alojohhardcore Thank you for this trade. It was a good call as always.

[Dominik Filkus @DominikFilkus]

@alojoh This is the way.

[Dominik Filkus @DominikFilkus]

@stamat18 @alojohhardcore Maybe this was the original post: https://t.co/KzYKmm0c0D

[Dominik Filkus @DominikFilkus]

@Dustin_Ryan_ @alojohhardcore I like this when its in the extreme fear territory

[Dominik Filkus @DominikFilkus]

@alojohhardcore @stamat18 Its a pro link. The post should be available without it i think.

[Dominik Filkus @DominikFilkus]

Even @t3dotchat has UX problems. This one loads forever. I do not know what is happening nowadays but seems everything is broken. Vibe coding ?

[Dominik Filkus @DominikFilkus]

@alojoh Instant buy

[Dominik Filkus @DominikFilkus]

@GoogleDeepMind Generations could be stuck too forever. Oh and Google Flow can't even work with mp3 files which are generated directly with Google Lyria 🤷‍♂️

[Dominik Filkus @DominikFilkus]

The @GoogleDeepMind Gemini Omni experience is so garbage that I can't even find words for it. Gemini app might be fine, but Flow is unusable.

[Claude @claudeai]

Introducing Claude Opus 4.8: it builds on Opus 4.7 with sharper judgment, more honesty about its own progress, and the ability to work independently for longer than its predecessors.

Available today at the same price.

[Serena Ge (Datacurve) @serenaa_ge]

Today we’re releasing DeepSWE, a new standard for agentic coding benchmarks.

On public leaderboards, top models often look relatively close in capability. DeepSWE shows where they actually diverge, reflecting the realistic experience of developers in their day-to-day work.