FastAPI + Claude Code sounds like a shortcut.But in this workflow, the agent doesn’t just get a prompt. It gets a project setup:✅ uv for the environment✅ Ruff for linting✅ ty for type checking✅ Project rules for direction✅ PyCharm for review and validation▶️ Watch the full video on building a FastAPI app with Claude Code, PyCharm, and Astral tools:
0:23
FastAPI + Claude Code sounds like a shortcut.But in this workflow, the agent doesn’t just get a prompt. It gets a project setup:✅ uv for the environment✅ Ruff for linting✅ ty for type checking✅ Project rules for direction✅ PyCharm for review and validation▶️ Watch the full video on building a FastAPI app with Claude Code, PyCharm, and Astral tools:
2.3K views1 month ago
x.comPyCharm, a JetBrains IDE
HERE’S HOW LOCAL LLM + HOME ASSISTANT KILLS GOOGLE HOMEstack: qwen3-30b-a3b-instruct + faster whisper large-v3-turbo + kokoro-fastapi, wyoming protocol as the glueall running on a single 3090 with 24gb vram at 275w, 29 smart devices connected, zero cloudask “do i need a jacket today” and it checks your weather sensor and actually answersfull breakdown in the article ↓
2:23
HERE’S HOW LOCAL LLM + HOME ASSISTANT KILLS GOOGLE HOMEstack: qwen3-30b-a3b-instruct + faster whisper large-v3-turbo + kokoro-fastapi, wyoming protocol as the glueall running on a single 3090 with 24gb vram at 275w, 29 smart devices connected, zero cloudask “do i need a jacket today” and it checks your weather sensor and actually answersfull breakdown in the article ↓
29.2K views3 weeks ago
x.comleopardracer
Round 1 Update
1:15
Round 1 Update
2.7K views5 hours ago
x.comSection 108
PixelForge 🫟Built entirely by Qwen 3.7 multi agent orchestration through Hermes on the Dgx SparkTwo AI users drawing simultaneously on a real time collaborative pixel canvas. The entire app was generated from a single prompt by Qwen 3.7- 6 AI agents across 3 parallel waves- 110,188 tokens generated in 128 seconds- 2,411 lines of code (921 server + 639 frontend + 851 tests)- FastAPI + WebSocket + rate limiting + structured error handling- 47/48 tests passing- Zero human coding
0:32
PixelForge 🫟Built entirely by Qwen 3.7 multi agent orchestration through Hermes on the Dgx SparkTwo AI users drawing simultaneously on a real time collaborative pixel canvas. The entire app was generated from a single prompt by Qwen 3.7- 6 AI agents across 3 parallel waves- 110,188 tokens generated in 128 seconds- 2,411 lines of code (921 server + 639 frontend + 851 tests)- FastAPI + WebSocket + rate limiting + structured error handling- 47/48 tests passing- Zero human coding
1.8K views1 month ago
x.comJoey
@lucas45588019 @treborfacts
0:59
@lucas45588019 @treborfacts
7.4K views3 days ago
x.com»Lu (a tortured poet)«
Post by ⋆｡˚ ❀ on X: 🫶🏻😭
0:05
Post by ⋆｡˚ ❀ on X: 🫶🏻😭
800 views13 hours ago
x.com⋆｡˚ ❀
Nipslip Fan / Instagram / Tiktok / Live / Clapper (@nipslipfan_fan) on X
0:05
Nipslip Fan / Instagram / Tiktok / Live / Clapper (@nipslipfan_fan) on X
7.9K views6 days ago
x.comNipslip Fan / Instagram / Tiktok / Live / Clapper
0:24
A SOLO DEV FROM TALLINN WALKED INTO A BANK WITH A $2,300 AMD BOX. BY FRIDAY, COMPLIANCE HAD SIGNED IT OFF AS A REPLACEMENT FOR THEIR $4,200-A-MONTH OPENAI CONTRACT11 clients now. 0 minutes of downtime this quarter. $0 in cloud API billsOrders a mini PC on the Ryzen AI Max+ 395. Installs Linux. Allocates 110GB as VRAM. Spins up Qwen3-235B locally. Writes a thin FastAPI wrapper. Boxes it upDrives to the client, slides it under the desk. Opens himself an SSH tunnel for updates. Pushes new model wei
2.2K views3 weeks ago
x.comchewa
0:04
kwon ohyul????
6.2K views1 day ago
x.comabel 🐈‍⬛
0:02
nyang maksudnye
309 views3 days ago
x.comcey
0:05
mbappe cuando caceres lo puteaba en guarani:
2.7K views1 week ago
x.comseo
0:35
Before your agent writes code, give it a constitution:✅ Mission✅ Tech stack✅ Roadmap✅ Rules for what comes nextThat’s spec-driven development – turning “build this” into a workflow the agent can follow and you can review.Watch the full video on #FastAPI + Claude Code + PyCharm, with spec-driven guardrails: ▶️
1.2K views1 month ago
x.comPyCharm, a JetBrains IDE
0:40
自己做了一个桌面端语音克隆工具，叫 tts-work。故事是这样的：小红书前阵子开源了 dots.tts，效果是真不错，但目前没有傻瓜式使用 app——你要会装 Python、装 PyTorch、跑命令行，对内容创作者来说门槛太高。现在做内容要配音，要么用 ElevenLabs 月费 $22，要么就对着麦克风一遍遍录。我想了一下，干脆自己撸一个。3 步用上：选一段参考音频（wav/mp3/flac/ogg/m4a 都行）告诉它参考音频里说了什么输入目标文本，点「合成」不用训练说话人模型，不用联网上传音频。最让我满意的是一个细节：输出 48kHz WAV。配音行业里这是"无损"档位，ElevenLabs 也只是 44.1kHz。技术栈——前端 Tauri 2.x + React + TypeScript（窗口比 Electron 轻 5 倍），后端 Python FastAPI + dots.tts 跑 WSL2 Ubuntu，PyTorch + CUDA + bfloat16 推理。Windows 下 start-all.bat 一键起前后端。最关键的：所有音频不传云端，不喂给任何大模
299 views3 weeks ago
x.comtiantian
1:24
Claude Codeで「毎回プロジェクト構造を説明しなきゃいけない」問題、たった1ファイルの配置で解消できる設計があります🐾よくある使い方は「各プロジェクトのルートにCLAUDE.mdを置く」ですよね。でも実は、documentsフォルダの**直下**に1枚だけ置く方法がもっと便利なんです。なぜかというと、Claude Codeには「起動時に親フォルダを上に遡ってCLAUDE.mdを探して読む」という仕様があります。つまりこういうことです：・documents/CLAUDE.md　← 全体の地図を書く・documents/project-A/・documents/project-B/・documents/project-C/project-Aを開いたとき、Claude Codeはproject-A内だけでなく、親のdocumentsフォルダのCLAUDE.mdも自動で読み込んでくれます。会社の就業規則（上位のCLAUDE.md）と部署マニュアル（プロジェクト固有のCLAUDE.md）が両方手元にある状態で動き出すイメージです。Before：プロジェクトを開くたびに「このリポジトリは
147 views2 weeks ago
x.comクロニキ｜Claude Code専門家
1:01
Para kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk
39.3K views11 hours ago
x.comMauro Triste com o Vasco da Silva
0:04
ELİM AYAĞIM TİTRİYO O NASIL HIRS O NASIL TWERKK KAVUŞTUKK
21.1K views1 day ago
x.compınos
0:12
화이띵하더니 강아지 인증🐶
1.7K views5 days ago
x.com설기
0:03
itu berdua gamau rangkulan ???
319 views4 days ago
x.comDeser
0:01
@SiiiJoker 😌😌😌 تعلم تعلم التشحليل 😁
115 views13 hours ago
x.comYahia
See more
Static thumbnail place holder