LogoThread Easy
  • 探索
  • 線程創作
LogoThread Easy

Twitter 線程的一站式夥伴

© 2025 Thread Easy All Rights Reserved.

探索

Newest first — browse tweet threads

Keep on to blur preview images; turn off to show them clearly

Q: what is, currently, a good production ready solution to "just run an AI 24/7 in a repository with a long term goal"? For example: "optimize HVM4 to achieve higher throughput" Just running Codex / Claude Code will result in the AI working for 1h max and then give up (they can't optimize past what it already is). But I'd gladly leave an agent working for several hours / days on the background if that means there is a slight chance of success. In particular, it needs to be something programmed to leave notes / accumulate memory, so it doesn't keep trying the same thing over and over. Solutions?

Q: what is, currently, a good production ready solution to "just run an AI 24/7 in a repository with a long term goal"? For example: "optimize HVM4 to achieve higher throughput" Just running Codex / Claude Code will result in the AI working for 1h max and then give up (they can't optimize past what it already is). But I'd gladly leave an agent working for several hours / days on the background if that means there is a slight chance of success. In particular, it needs to be something programmed to leave notes / accumulate memory, so it doesn't keep trying the same thing over and over. Solutions?

Kind / Bend / HVM / INets / λCalculus

avatar for Taelin
Taelin
Fri Dec 19 10:14:51
Q: what is, currently, a good production ready solution to "just run an AI 24/7 in a repository with a long term goal"? For example: "optimize HVM4 to achieve higher throughput" Just running Codex / Claude Code will result in the AI working for 1h max and then give up (they can't optimize past what it already is). But I'd gladly leave an agent working for several hours / days on the background if that means there is a slight chance of success. In particular, it needs to be something programmed to leave notes / accumulate memory, so it doesn't keep trying the same thing over and over. Solutions?

Q: what is, currently, a good production ready solution to "just run an AI 24/7 in a repository with a long term goal"? For example: "optimize HVM4 to achieve higher throughput" Just running Codex / Claude Code will result in the AI working for 1h max and then give up (they can't optimize past what it already is). But I'd gladly leave an agent working for several hours / days on the background if that means there is a slight chance of success. In particular, it needs to be something programmed to leave notes / accumulate memory, so it doesn't keep trying the same thing over and over. Solutions?

Kind / Bend / HVM / INets / λCalculus

avatar for Taelin
Taelin
Fri Dec 19 10:14:51
Q: what is, currently, a good production ready solution to "just run an AI 24/7 in a repository with a long term goal"? For example: "optimize HVM4 to achieve higher throughput" Just running Codex / Claude Code will result in the AI working for 1h max and then give up (they can't optimize past what it already is). But I'd gladly leave an agent working for several hours / days on the background if that means there is a slight chance of success. In particular, it needs to be something programmed to leave notes / accumulate memory, so it doesn't keep trying the same thing over and over. Solutions?

Q: what is, currently, a good production ready solution to "just run an AI 24/7 in a repository with a long term goal"? For example: "optimize HVM4 to achieve higher throughput" Just running Codex / Claude Code will result in the AI working for 1h max and then give up (they can't optimize past what it already is). But I'd gladly leave an agent working for several hours / days on the background if that means there is a slight chance of success. In particular, it needs to be something programmed to leave notes / accumulate memory, so it doesn't keep trying the same thing over and over. Solutions?

Kind / Bend / HVM / INets / λCalculus

avatar for Taelin
Taelin
Fri Dec 19 10:14:51
I’m super happy how diverse my current product portfolio is: PDF Vector: B2B ($X MRR) [Stealth project]: B2C ($XX MRR) @Vibingbase: B2B & B2C (potential 🚀) @HackerResidency: Community (invaluable ❤️) For the first time, I don’t feel the urge to build new ideas anymore 🥹

I’m super happy how diverse my current product portfolio is: PDF Vector: B2B ($X MRR) [Stealth project]: B2C ($XX MRR) @Vibingbase: B2B & B2C (potential 🚀) @HackerResidency: Community (invaluable ❤️) For the first time, I don’t feel the urge to build new ideas anymore 🥹

Built 8 startups in 12 months • Sold 3/8 startups • Building @HackerResidency, https://t.co/ns7SH9FiWh, https://t.co/1tAr7flVT2, https://t.co/sYr4JzzOtJ, https://t.co/HvvzdvWg5H, a secret startup 🤫

avatar for Minh-Phuc Tran
Minh-Phuc Tran
Fri Dec 19 10:02:17
如果你想在自己的电脑或手机上搞一个速度快、不联网、不花钱的 AI 模型。 那可以去试下 Nexa SDK 这个开源项目。 它可以说是目前市面上集成度最高、工具链最完整的开源方案之一。 GitHub：https://t.co/WVR0UnPhuh 靠着自研的模型压缩技术，即便完全断网，它也能在几毫秒内搜遍本地几千张相片。 处理上万份敏感文档也不在话下，数据始终留在本地，安全性与隐私性远超云端模型 API。 无缝兼容 OpenAI API 接口，方便我们直接集成到现有的开发工作流当中。 另外，它还提供开箱即用的命令行工具，一行代码就能完成模型的下载与运行，上手特别简单。

如果你想在自己的电脑或手机上搞一个速度快、不联网、不花钱的 AI 模型。 那可以去试下 Nexa SDK 这个开源项目。 它可以说是目前市面上集成度最高、工具链最完整的开源方案之一。 GitHub：https://t.co/WVR0UnPhuh 靠着自研的模型压缩技术，即便完全断网，它也能在几毫秒内搜遍本地几千张相片。 处理上万份敏感文档也不在话下，数据始终留在本地，安全性与隐私性远超云端模型 API。 无缝兼容 OpenAI API 接口，方便我们直接集成到现有的开发工作流当中。 另外，它还提供开箱即用的命令行工具，一行代码就能完成模型的下载与运行，上手特别简单。

💡 挖掘开源的价值 🧑🏻‍💻 坚持分享 GitHub 上高质量、有趣、实用的教程、AI工具、前沿 AI 技术 🧐 A list cool, interesting projects of GitHub. ✏️ 公众号：GitHubDaily

avatar for GitHubDaily
GitHubDaily
Fri Dec 19 10:00:04
Only 10 years ago: - El Salvador 🇸🇻 had the worlds highest homicide rate (now the lowest) - Peru 🇵🇪 passport holders had visa free access to 85 countries (now 150) - UK 🇬🇧 had non-dom tax laws - Cyprus 🇨🇾 still sold their passports A lot can change (for better or worse)

Only 10 years ago: - El Salvador 🇸🇻 had the worlds highest homicide rate (now the lowest) - Peru 🇵🇪 passport holders had visa free access to 85 countries (now 150) - UK 🇬🇧 had non-dom tax laws - Cyprus 🇨🇾 still sold their passports A lot can change (for better or worse)

🌏 RCBI advisor & offshore services for HNWI, business owners with a focus on:🇨🇭🇲🇾🇰🇳🇵🇾🇻🇺🇳🇷🇵🇦🇱🇻🇦🇪🇭🇰 | Geopolitics | Healthy lifestyle

avatar for Lord Y. Fouzi 🇲🇾🇨🇭
Lord Y. Fouzi 🇲🇾🇨🇭
Fri Dec 19 09:57:27
  • Previous
  • 1
  • More pages
  • 346
  • 347
  • 348
  • More pages
  • 5634
  • Next