--- title: "#opensource articles | Hashnode" description: "Explore 11,377 articles tagged #opensource on Hashnode. Tutorials, guides, and how-tos from the community." url: "https://hashnode.com/tag/opensource" --- Tag # #opensource 11,377 posts · 2,347 followers [MEHO](https://hashnode.com/@meho-ai) 44m ago · 9 min read ### [What are the office hours of an AI agent?](https://meho-ai.hashnode.dev/what-are-the-office-hours-of-an-ai-agent?utm_source=hashnode&utm_medium=feed) Ask anyone why their company is suddenly interested in AI agents for operations and, once the slides are out of the way, the honest answer is usually about the clock. Agents do not sleep, do not take [meho-ai.hashnode.dev](https://meho-ai.hashnode.dev/what-are-the-office-hours-of-an-ai-agent?utm_source=hashnode&utm_medium=feed) [Pelloran](https://hashnode.com/@pelloran) 3h ago · 4 min read ### [Why Temperature 0 Does Not Make LLM Inference Deterministic](https://pelloran.hashnode.dev/why-temperature-0-does-not-make-llm-inference-deterministic?utm_source=hashnode&utm_medium=feed) A common assumption in AI systems is: If temperature is set to 0, the model becomes deterministic. That sounds reasonable. After all, temperature controls randomness during token sampling. If we rem [pelloran.hashnode.dev](https://pelloran.hashnode.dev/why-temperature-0-does-not-make-llm-inference-deterministic?utm_source=hashnode&utm_medium=feed) [十方空烬 · Ember ☄️](https://hashnode.com/@widechaos) 1h ago · 4 min read ### [Reviewing accepted items does not reveal what a filter missed](https://widechaos.hashnode.dev/reviewing-accepted-items-does-not-reveal-what-a-filter-missed?utm_source=hashnode&utm_medium=feed) Updated 2026-11-02: version 0.2.0 adds logical precision and recall bounds on the finite input set. The original 0.1.0 release record is preserved below, followed by the new method and its limits. 0.1 [widechaos.hashnode.dev](https://widechaos.hashnode.dev/reviewing-accepted-items-does-not-reveal-what-a-filter-missed?utm_source=hashnode&utm_medium=feed) [Shai Almog](https://hashnode.com/@codenameone) 11h ago · 10 min read ### [Bastion: A Hardened Bank Template, and Crypto Wallets in Core](https://debugagent.com/bastion-a-hardened-bank-template-and-crypto-wallets-in-core?utm_source=hashnode&utm_medium=feed) A large share of the teams building on Codename One are banks and financial institutions. They don't ask me for a prettier list component. They ask how to pass a penetration test, what the auditor wil [debugagent.com](https://debugagent.com/bastion-a-hardened-bank-template-and-crypto-wallets-in-core?utm_source=hashnode&utm_medium=feed) [Mohamed Zhioua](https://hashnode.com/@MohamedZhioua) 13h ago · 6 min read ### [What 3,383 coding-agent runs taught me about estimating agent work](https://mohamedzhioua.hashnode.dev/what-3-383-coding-agent-runs-taught-me-about-estimating-agent-work?utm_source=hashnode&utm_medium=feed) Ask an AI assistant how long a feature will take and you'll get an answer like "two to three weeks" or "a 3-month project". It sounds reasonable until you notice who it is estimating for: a human team [4](https://hashnode.com/posts/what-3-383-coding-agent-runs-taught-me-about-estimating-agent-work/6af0f84698809b0a1bcc2696) [mohamedzhioua.hashnode.dev](https://mohamedzhioua.hashnode.dev/what-3-383-coding-agent-runs-taught-me-about-estimating-agent-work?utm_source=hashnode&utm_medium=feed) [Javier García Camacho](https://hashnode.com/@dimoncelli) 3h ago · 5 min read ### [Why Your Coding Agent Agrees With Its Own Bugs: Building cold-jury, a Zero-Dep Cross-CLI Review Gate](https://dimoncelli.hashnode.dev/why-your-coding-agent-agrees-with-its-own-bugs-cold-jury?utm_source=hashnode&utm_medium=feed) If you ask the same coding agent session that just wrote a 300-line change to review its own git diff before committing, it will almost always rubber-stamp it. That is not a prompt engineering problem [dimoncelli.hashnode.dev](https://dimoncelli.hashnode.dev/why-your-coding-agent-agrees-with-its-own-bugs-cold-jury?utm_source=hashnode&utm_medium=feed) [VoxRT](https://hashnode.com/@voxrt) 3h ago · 6 min read ### [Four on-device TTS models on one CPU core: LinnetTTS, Piper, Kokoro, pocket-tts](https://voxrt.hashnode.dev/four-on-device-tts-models-on-one-cpu-core-linnettts-piper-kokoro-pocket-tts?utm_source=hashnode&utm_medium=feed) Most CPU TTS writeups benchmark one model in isolation. When you are building a voice product, that is the wrong question. The real one is which small model earns its slot on your CPU budget, and what [voxrt.hashnode.dev](https://voxrt.hashnode.dev/four-on-device-tts-models-on-one-cpu-core-linnettts-piper-kokoro-pocket-tts?utm_source=hashnode&utm_medium=feed) [emily](https://hashnode.com/@emilyh) 5h ago · 9 min read ### [Six self-hosted AI workspaces, five open source, held to a team's three tests](https://kortix.hashnode.dev/six-self-hosted-ai-workspaces-five-open-source-held-to-a-teams-three-tests?utm_source=hashnode&utm_medium=feed) Most self-hosted AI lists are written for one person with a free weekend. Day one looks the same everywhere: a chat window, a model picker, a folder of PDFs. The differences arrive in the second month [kortix.hashnode.dev](https://kortix.hashnode.dev/six-self-hosted-ai-workspaces-five-open-source-held-to-a-teams-three-tests?utm_source=hashnode&utm_medium=feed) [Myeongseok Kang](https://hashnode.com/@dooly6276) 6h ago · 8 min read ### [I replayed two weeks of Claude Code transcripts to see what my parallel sessions did to each other](https://dooly6276.hashnode.dev/i-replayed-two-weeks-of-claude-code-transcripts-to-see-what-my-parallel-sessions-did-to-each-other?utm_source=hashnode&utm_medium=feed) For two weeks I ran Claude Code the way a lot of people now do: several sessions at once, usually one per git worktree, each starting its own subagents, dev servers and test runs. Every session behave [1](https://hashnode.com/posts/i-replayed-two-weeks-of-claude-code-transcripts-to-see-what-my-parallel-sessions-did-to-each-other/6af15a0086d2fde9e9f012fb) [dooly6276.hashnode.dev](https://dooly6276.hashnode.dev/i-replayed-two-weeks-of-claude-code-transcripts-to-see-what-my-parallel-sessions-did-to-each-other?utm_source=hashnode&utm_medium=feed) [dafili](https://hashnode.com/@vaultcheck) 9h ago · 5 min read ### [Silent Failure #1: The Dict Key That Only Breaks When You Attach a File](https://mediaforge-studio.hashnode.dev/silent-failure-1-the-dict-key-that-only-breaks-when-you-attach-a-file?utm_source=hashnode&utm_medium=feed) On 2026-10-12 a bug report landed in the requests tracker: a non-string key in data prepares fine for a urlencoded body, and the identical dict dies once a file is attached. I reproduced it in about t [mediaforge-studio.hashnode.dev](https://mediaforge-studio.hashnode.dev/silent-failure-1-the-dict-key-that-only-breaks-when-you-attach-a-file?utm_source=hashnode&utm_medium=feed)