The smart news app - AI RSS feed
Your news feed.
Your algorithm.
Every other algorithm is a black box built for ads and outrage.
Here, the algorithm is yours — it actually filters for what you want to read.
Build your own information system
Your information system, end to end: the sources you picked, an algorithm that is yours alone, your AI agent filtering again on top, delivered the way you like to read.
Every source you care about, in one place.
- Paste a URL — the feed is auto-discovered
- Substack · podcasts · Bluesky · Google News
- OPML import & export, so you’re never locked in
Signal, not noise
A daily briefing synthesized across every source
Serving LLMs got ~10× cheaper — it reshuffles who can compete.
Rate-cut odds jumped; risk assets caught a bid across your feeds.
Trending radar catches it before it peaks
Track what matters — a ticker, a topic, a rival
One account, many personas
Your work self, your investor self, your side-obsession self — in Feedpik each is a persona with its own RSS feeds and ranking.
The Feedpik Persona Community
A Persona is a recipe: hand-picked RSS feeds plus a tuned ranking policy. Install one and your Feedpik feed is instantly expert-grade. Publish yours and build a following.
Central banks and the earnings that move indices. Primary sources over hot takes.
Platform shifts and audience growth. Long essays over engagement bait.
Model releases and the papers behind them. Research over launch-day hype.
Protocol upgrades, on-chain data, and regulation. Research over influencer noise.
AI that reads alongside you
- · Calibration can reuse error-correction data — no separate pass.
- · Removes qubit drift as a blocker for long algorithms.
AI summaries
One tap turns a long read into its key points — the gist in seconds, not minutes.
…whether we can make enough hardware qubits to connect into the error-corrected logical qubits we need, and how we generate the states for universal computation…
Groups of physical qubits bundled with error correction so they behave as one reliable qubit.
AI explanations
Hit a term, a chart, or a bit of jargon? Get it explained in the article’s own context.
AI Q&A
Ask a question about the piece and get an answer grounded in the actual text — without leaving the page.
Two shifts landed this week that point the same direction: the cost of…
Reading → draft
Pull the articles you care about into a workspace and generate a first draft from your sources.
Fair questions
Is this just another RSS feed reader?+
Under the hood Feedpik speaks RSS, Atom, and OPML — so you can pull in any source on the open web. But the point is what happens after: it works as an AI RSS feed reader, ranking everything for you and letting you (or your agent) control the ranking. Most readers show a raw timeline or a black box; Feedpik is an RSS feed you actually own.
How do I add an RSS feed?+
Paste the website address — Feedpik finds the RSS feed for you. If a site publishes several feeds you pick the one you want, and a direct RSS or Atom feed URL works too. Substack publications, podcasts, YouTube channels, Bluesky accounts, and Google News searches all resolve to a feed automatically.
Can I import the RSS feeds I already follow?+
Yes — export an OPML file from Feedly, Inoreader, NetNewsWire, or any other RSS reader and import it in one step. Feedpik exports OPML too, so your subscriptions are never trapped here either.
Will my reading history be shared when I publish a Persona?+
No — in Feedpik that is structurally impossible, not just a policy promise. A published Persona contains only RSS feeds and tuning parameters (weights, filters, prompts). Your reading behavior and interest vectors never leave your account.
Can I take my data with me?+
Yes. Your RSS feeds export from Feedpik as standard OPML, ranking policies export as JSON, and your behavior data is yours to view, edit, and delete. No lock-in by design.
How does my AI agent connect?+
Create a scoped Feedpik API key (read-only up to policy-tuning permissions — you choose). Any agent that speaks HTTP can pull candidates, record feedback, and propose policy changes. Every agent change creates a new version with an audit trail.
Finally, your data and your algorithm belong to you.
Feedpik is private by structure, not by promise