Your Bookmarks Remember You

You saved that article for a reason. Maybe it was a brilliant framework, a quote that resonated, or research for a project that never happened. Now it’s buried under hundreds of other bookmarks, forgotten.

Memory Jogger changes that.

Every day, Arivu surfaces bookmarks you haven’t visited in a while—but that connect to what you’re working on now. It’s not random. Our AI scores connections between your bookmarks based on semantic relationships and your recent activity, surfacing the ones that matter most.

The average knowledge worker saves hundreds of bookmarks per year. Studies show that most are never revisited after the first week. That’s not a personal failing—it’s a tool problem. Traditional bookmarks are passive storage. Memory Jogger turns your collection into an active collaborator.

How It Works

When you open your dashboard, Memory Jogger analyzes your recent saves and searches, then scans your entire collection for related content you might have forgotten.

That article on system design from two years ago? It suddenly becomes relevant when you’re researching architecture patterns today.

The AI doesn’t just match keywords—it understands concepts. A bookmark about “decision fatigue” might surface when you’re saving articles about productivity. An old piece on narrative structure might reappear when you’re researching storytelling in product design. These aren’t coincidences. They’re semantic connections your brain would make if it could hold your entire collection in working memory.

The Scoring Engine

Behind the scenes, Memory Jogger runs a multi-factor scoring algorithm on your collection. Each bookmark is evaluated against your recent activity across four dimensions:

Semantic similarity. Using embeddings from Gemini 2.5 Flash, we calculate how conceptually close each old bookmark is to your recent saves and searches. A 92% similarity between “cognitive load in UI design” and your new save about “reducing user friction” isn’t keyword matching—it’s meaning matching.

Temporal decay. Bookmarks you haven’t touched in months get a boost. The goal is resurfacing, not reinforcing what you already remember. A brilliant article from 18 months ago that’s still relevant today is exactly what Memory Jogger is designed to find.

Engagement signals. Bookmarks you’ve previously clicked from Memory Jogger suggestions get weighted differently than those you’ve consistently skipped. The system learns your preferences without requiring explicit feedback.

Context clustering. If you’ve been saving a lot of content around a specific topic this week, Memory Jogger looks for older bookmarks that might belong to that same cluster—even if you didn’t tag them that way originally.

Why This Matters

Most “rediscovery” features in other tools are either random or based on crude heuristics like “bookmarks from this day last year.” That’s nostalgia, not utility.

Memory Jogger is built around a simple principle: relevance is contextual. What’s useful today depends on what you’re working on today. The same bookmark might be irrelevant for six months, then suddenly become exactly what you need when your focus shifts.

We’ve seen users rediscover research they forgot they’d saved, reconnect ideas across projects, and find citations for work-in-progress that would have taken hours to re-research. That’s the promise: your past saves become present insights.

Right Where You Need It

You’ll find Memory Jogger in your dashboard sidebar. No extra clicks, no separate page to visit. Just a gentle daily reminder of the knowledge you’ve already collected.

Each suggestion shows you why it was surfaced, so you can quickly decide if it’s worth revisiting or skip to the next one. The explanations are specific: “Similar to your recent save about API design patterns” or “Connects to your ‘product research’ collection.” No vague “You might like this” messages.

Designed for Flow

We deliberately placed Memory Jogger in the sidebar rather than as a separate feature because context switching kills productivity. When you’re in research mode, the last thing you need is to navigate away from your current view to check suggestions.

The sidebar integration means you can glance at today’s suggestions while reviewing your recent saves. See something relevant? Click to open. Not interested? Keep scrolling. The suggestions refresh daily, so there’s never pressure to act immediately.

Transparent Reasoning

Every suggestion includes a brief explanation of why it was surfaced. This isn’t just a nice-to-have—it’s essential for building trust in AI-powered features.

When you understand why a two-year-old article about database indexing appeared alongside your recent saves about query optimization, you’re more likely to click through. And when you understand the connection, you’re more likely to make the creative leap the system was designed to enable.

Think of it as your knowledge collection gently tapping you on the shoulder: “Hey, remember this? It might help with what you’re working on.”

Other Changes

• Arivu Blog launch — Deep-dive guides for knowledge workers are now live. First up: how to build a personal knowledge system that actually works, because saving bookmarks is easy—finding them again shouldn’t be the hard part. We’re publishing new guides weekly on topics ranging from research workflows to building a second brain.
• Dashboard UI refactor — Memory Jogger now lives in the sidebar with a cleaner, more integrated design. The new layout puts rediscovery suggestions front and center without cluttering your main bookmark view. We also improved the visual hierarchy to make scanning faster and reduced visual noise across the interface.
• Backend improvements — We rebuilt the AI-powered connections scoring engine for faster, more accurate suggestions. Memory Jogger now processes your collection more efficiently and surfaces better matches. Average scoring time dropped from 800ms to under 200ms, meaning suggestions load instantly when you open your dashboard.