AI on Arivu

Your Second Brain, Your Model Provider

Sat, 11 Jul 2026 00:00:00 +0000

The model connected to your second brain should be your choice. Arivu now supports Gemini, Anthropic, OpenAI-compatible services, and local or custom-compatible endpoints through runtime Admin Settings.

You can change the provider without rebuilding the app or editing deployment files. Admins choose a preset or compatible endpoint, select the model, and add credentials when that provider requires them. Local services can be keyless where supported.

One Workflow, Different AI Layers

Hosted providers can offer stronger generated summaries and synthesis. A local endpoint can keep more processing inside infrastructure you control. An OpenAI-compatible endpoint can connect another service that follows the expected interface. The right choice depends on your privacy, quality, cost, and operational needs.

An AI Assistant You Can Actually Trust

Sun, 05 Jul 2026 00:00:00 +0000

Update, July 11, 2026: Arivu now supports runtime choice among Gemini, Anthropic, OpenAI-compatible, local, and custom-compatible providers. Local endpoints can be keyless where supported, and AI remains optional. Read Your Second Brain, Your Model Provider for the current model-provider story.

Most AI assistants ask you for trust up front. They act, and you find out afterward. That is a bad fit for a tool that holds your knowledge and can change it.

Your X Bookmarks, Finally Useful

Sun, 08 Feb 2026 00:00:00 +0000

Historical note, July 2026: X bookmark sync remains optional and depends on X API access. X text is authoritative evidence for the post, while a linked article is separate evidence used only after successful extraction. Metadata-only or failed extraction does not support generated claims, and unsupported derived material stays out of Graph and analytical Insights. Imported items otherwise remain in the broader second-brain loop, and AI enrichment is provider-optional.

You bookmark tweets. A lot of them. A thread on system design, a link to a blog post about pricing strategy, someone’s take on a paper you want to read later. They pile up in X’s bookmarks tab: a flat, unsearchable list with no summaries, no tags, no way to connect ideas.

Search That Thinks, Summaries That Matter

Thu, 05 Feb 2026 00:00:00 +0000

Historical note, July 2026: This post describes an earlier search and summary milestone. Search, summaries, and source-aware answers still matter, but the current product frames them inside a broader self-hosted second-brain workflow with notes, Inbox, Focus, Review, optional AI, and cited answers.

Finding a bookmark shouldn’t feel like searching for a needle in a haystack. And reading an AI summary shouldn’t feel like reading filler text generated to hit a word count.

Your Bookmarks Remember You

Fri, 16 Jan 2026 00:00:00 +0000

Historical note, July 2026: Memory Jogger was the first expression of Arivu’s rediscovery idea. That idea now lives inside the broader Review loop, which can bring back bookmarks, imported document notes, standalone notes, and related research context because they are due, older, important, unfinished, or still actionable, and lets you give recall feedback such as Useful, Not useful, Snooze longer, or Never resurface. See Review Learns What Matters and Research Has More Shape Now for the current model.

Your Brain, Mapped

Thu, 08 Jan 2026 00:00:00 +0000

Historical note, July 2026: This post captures an early step beyond folders and tags. The Knowledge Graph still matters, but it now sits inside a broader second-brain workflow: notes, explicit links, backlinks, tasks, reminders, cited answers, and Review. Older references to analytics, scoring, and AI-only processing should be read as historical context, not the whole current product direction.

Today, Arivu stops being a bookmark manager.

For years, bookmarking tools have asked the same question: where do you want to put this? Folders. Tags. Collections. We organize our knowledge like filing cabinets because that’s what computers understood.