paddy (@0xPD33) header
paddy (@0xPD33) profile picture

schizo tech wizard. 🧙‍♂️ Christ is King ✝️ NixOS ❄️ | AI/ML | Game Dev | DeFi

heaven on earth
Joined May 2023
Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)Photo by paddy (@0xPD33)
Pinned Tweet
I haven't written a single line of code in 1 year. 2 of the projects I was working on are still alive in a way. I decided to continue building out the FPS game and rewrite the Godot Open World in a custom C + Vulkan engine. There are no limits to what you can do in game development at this point, the latest closed source and open source models are good enough for everything you can imagine.
I haven't written a single line of code in months. And yet, I was able to start building 4 large-scale games. 7 years in game dev. Here's what's actually different now. The projects, all side projects, all built with AI over the past few months: → Open world RPG prototype (GDScript/Godot) → Space sim in Godot-Rust (Rimworld × Starsector) → FPS in C + SDL3 → Life-sim engine in C + SDL3, think Sims 2/3 but ultra-scalable with planned LLM integration Here is what the leading models handled without issues: - Custom render pipelines on top of SDL - Shaders for everything I could describe or give references for - Procedural world and model generation (everything you see in the screenshots is procedural, no external assets) - Complex, interlocked gameplay systems Years of programming work condensed into weeks. The engine doesn't matter anymore. Godot and GDScript, Rust and Bevy, C and SDL/Raylib, Javascript and ThreeJS. Pick whatever fits. The models handle it. What actually matters is domain knowledge. 7 YOE meant I knew what to ask for, how to structure things, which patterns to reach for. That's the real multiplier. There are things these models still can't do well. I'll break down each project with the specific challenges and what I'm planning in future posts. But one thing is clear: the floor has shifted. If you can clearly describe your dream game, you can almost certainly start building it today. Fire up Claude or Codex and give it a shot. Any questions? Feel free to ask.
7
2,217
Just used a Claude Fable reset and it used two at once wtf? I got no reset left for Fable now. Went straight from 2 to 0. @AnthropicAI helpppp
3
152
@bcherny @ClaudeDevs @lydiahallie @alexalbert__ any way I can get some help on this? my main goal is not even getting the extra reset back (its a shame but not the end of the world) but perhaps this is a bug that affects other users. I don't have screenshots of before the extra limit was erased either but I'm willing to share more info in DM's
26
Replying to @0xPD33 @AnthropicAI
No Fin that was not what I was looking for. It's over.
33
I remember trying to vibe code stuff with GPT-3.5-Turbo and GPT-4. Already saw the potential back then but the output didn't "just work" most of the time because even a request to change a single file and write some code came with hallucinations. The nostalgia is already kicking in for me because it was also kinda fun (but also tiresome) to work with a worse model and correct its mistakes.
Remember when ChatGPT had 25 GPT-4 messages every 3 hours? The good old times
2
81
Fable 5.2 is really amazing so far. Makes sense to give the big boy tasks and designs to it and make it use Opus/Sonnet subagents for lower intelligence stuff.
1
1
106
Today an LLM hallucination resulted in 72 eggs being delivered to my house. 🥚🥚🥚🥚
44
712
7
14,880
356,845
paddy retweeted
opsec level: just give up
your HDMI cable can leak what’s on your screen watch the laptop on the left. it reconstructs the video playing on the monitor beside it from radio signals leaking out of the HDMI connection this is TempestSDR. an antenna picks up the leakage, a HackRF radio captures it, and software turns it back into an image researchers took this further with Deep-TEMPEST, using AI to make recovered images and text easier to read. this clip shows the original technique quality depends on shielding, distance and interference. the picture is imperfect, but recognizable content still gets through a wired connection can still leave a wireless trace
Readers added context they thought people might want to know
HDMI TEMPEST demos like TempestSDR recover distorted images only in controlled close-range setups often using unshielded or poorly shielded cables; standard consumer HDMI connections and distances prevent practical screen leakage or remote recovery. rtl-sdr.com/easvesdropping… arxiv.org/pdf/2407.09717 techradar.com/computing/cybe… kaspersky.com/blog/deep-temp…
145
2,069
35
41,689
702,293
You have to relentlessly push your agents and LLMs to go into the right direction. Leave to much to chance and it starts fucking up or starts making obviously wrong choices in your code. This makes or breaks projects imo
1
4
267
I have a confession: I don't actually like phones. At heart I'm a game and engine dev, I like DeFi, I enjoy tinkering with Linux. I'm the furthest from a mobile dev. Before LLMs were great at frontend I absolutely hated doing responsive design. I also think people should return to the desktop as I've said in the past... But since most of the population uses phones, I might as well try some things. @askcommonplace will continue and it's just the start. Software has never been easier and there is a real opportunity for making everything better, including what runs on phones. I'm also thinking about the distribution aspect. F-Droid is good, GitHub releases work. Play Store/App store? Not a fan. What is the future of distribution for mobile apps if Google enforces ID checks and all that? Maybe this is where crypto and DeFi can be useful. If you're thinking about this too, let's chat.
1
9
346
Show me one engineer who writes better code than an llm.
58
543
32
4,696
75,444
Just a reminder that Google announced that all Android app developers must register centrally, pay a fee, and submit government ID, or their apps will be blocked on every device. I won't comply. keepandroidopen.org @AlteredDeal #KeepAndroidOpen
1
2
214
If they really end up doing this, the best solution is probably GrapheneOS or a Linux distro on your phone. Invest in your privacy. Get compute for local AI and a phone that supports GrapheneOS.
1
1
74
Blender × Claude Opus 5.5テスト かなり素晴らしいクオリティ tripoも使用許可したけど、5%くらいしか使ってなくて、ほぼ自分でモデリングしてた 文字など一部のテクスチャにChatGPT使いつつ、シェーダーもほぼプロシージャルに仕上げてくれてる 音はHunyuanVideo-Foley メイキングも作ってもらった
82
441
65
5,315
367,689
paddy retweeted
May you be blessed with hardship and tough times such that you may grow and prove yourself worthy.
247
693
90
8,049
208,308
custom tailored feeds are necessary at this point, can't be rawdogging certain social media platforms anymore #stoptheslop
1
5
173
if it really supports GrapheneOS I just found my next phone purchase
Some standards are met. Others are set. A unique expression of design. Introducing motorola signature 27. Designed to a higher standard. Powered by @Snapdragon's most advanced chipset yet, #Snapdragon 8 Elite Extreme Gen 6. #MotorolaSignature27 #SignatureStatement Learn more: bit.ly/48y042D
3
169
Took me a while but here we are. Confident now that my app is among the fastest of its kind on Android, but there is still a lot of potential.
Commonplace 0.4.0 is out. An offline research app for Android that answers from a library on your phone and cites its sources. Now it's way faster, and it reads answers aloud. No account. No cloud. No network permission at all.
1
2
9
400
paddy retweeted
Asked one of our developers how he uses his AI agents and he just sent me this:
179
1,337
270
16,207
1,775,860
5
19
588
8,039