S

Sprinklz

Free RSS reader with a recommendation engine you can configure and rewrite yourself

Free
4.6 (8 reviews)

Gallery

About Sprinklz

Sprinklz is a free news and RSS reader built around a recommendation engine you control rather than one that controls you. It does what any reader does, pulling in blogs, news sites, and YouTube channels, then adds a layer most readers don't have, with real machine learning and NLP tools you can tune and the ability to open the ranking source code and edit it directly from the browser.

The reasoning behind it is stated plainly on the site. Its creator argues that algorithmic bias is destroying everything we know and love about modern media, because mainstream algorithms push the most aggravating and addictive content to keep you scrolling as long as possible. He describes losing two hours a day to social media and watching every attempt to fix his feed get nudged back toward emptier material, where a video about emacs leads to editor-war drama, then hollow edutainment, then general-purpose clickbait. He points out that you can't even search for something without recommended distractions crowding in. His stated goal is bringing back digital autonomy, so people can make their own choices about what they see and use digital media in a way that meaningfully improves their lives rather than functioning as a necessary evil.

In practice that means a reader with sorting and filtering, a built-in search engine for discovering feeds, and NLP features like topic similarity and clustering, so related stories group together and you can steer coverage by subject instead of by source. Themes and styles are deeply customizable, and the site claims pretty much everything on it is modifiable. The headline feature is source-code access. Instead of picking from a handful of preset ranking options, you open the algorithm and change how it works, which is a different proposition from the sliders and mute lists most readers offer. The stated aim is giving you tools with the power of real machine learning while keeping you in charge of how they behave.

You can try it before signing up, which is a good sign for something this configurable. Six demos run straight from the homepage. Top Headlines pulls major stories from the biggest newspapers. Hackable Hacker News is an implementation of Hacker News where you edit the ranking algorithm yourself, and it's the clearest demonstration of the whole idea. Another demo covers following your favorite blogs, websites, and YouTube channels in one place. One opens the source code directly in the browser, on the logic that direct access means you can do whatever you want. A fifth combines multimodal data, splicing sources of different types according to their relative importance at their origin. The last covers building custom themes.

It fits people who still miss Google Reader and want more than a chronological list back, and especially developers who'd rather write a scoring function than accept somebody else's defaults. If you want a reader that works out of the box with zero configuration, the appeal is thinner, because the knobs are the entire point. The audience is people who treat tuning their own feed as a feature rather than a chore, and who care enough about what reaches them to spend an evening getting it right. For everyone else, a simpler reader will be less work.

What separates it from the field is where the control sits. Most readers give you folders, tags, and maybe a mute list, and a few bolt on opaque AI recommendations you can't inspect or correct when they get it wrong. Sprinklz treats ranking logic as something you're meant to open and rewrite, which is a genuinely different bargain. You take on the work of shaping the feed, and in exchange you know exactly why any given story showed up, because you wrote the rule that put it there. That traceability is hard to get anywhere else without building your own reader from scratch.

It's free to use, with accounts through a standard signup and no pricing page anywhere on the site. The project comes from a solo developer named Sammy, who spent five years as a software engineer at Google before leaving to take time off and work on projects, and who has been building this one for roughly three years. He describes it as his big project and says he's hoping to give back and do some good in the world. Community and support run through a public Discord linked from the site. Worth knowing that it's an independent effort with no published pricing and no documented roadmap, so it's best treated as a promising indie reader rather than a company-backed platform with support guarantees.

Key Features

  • Configurable recommendation engine
  • In-browser source code editing
  • Built-in search engine for feeds
  • Topic similarity and clustering
  • Sorting and filtering controls
  • Fully customizable themes and styles

Pros & Cons

What we like

  • Free to use with demos you can try before signing up
  • Ranking logic is open to inspect and rewrite
  • NLP clustering groups related stories by topic
  • Follows blogs, news sites, and YouTube channels in one place

Room for improvement

  • Independent project with no published pricing
  • Getting real value takes more setup than a default feed
  • Small community, with support running through Discord
  • Source-code editing appeals mainly to developers

Frequently Asked Questions

What is Sprinklz?
Sprinklz is a free news and RSS reader with a configurable recommendation engine. It has the sorting, filtering, and feed search you'd expect, plus NLP tools like topic similarity and clustering, and it lets you edit the ranking source code directly from the browser.
Is Sprinklz free?
Yes. It's free to use and there's no pricing page on the site. You can try six demos from the homepage without an account, and create a login when you want to save your own feeds. It's an independent project, so pricing could change as it matures.
How is Sprinklz different from other RSS readers?
Most readers give you folders and tags, and some add recommendations you can't inspect. Sprinklz exposes the ranking algorithm as source code you can open and rewrite, and pairs that with machine learning features like clustering. The trade is more control in exchange for more configuration.
Who is Sprinklz for?
Readers who want to decide what reaches them instead of letting an engagement algorithm decide, and developers comfortable tweaking a scoring function. It's built by Sammy, a former Google software engineer who has spent roughly three years on it, with community and support in a public Discord.

Best For

Building a news feed without algorithmic nudgingRewriting a Hacker News ranking algorithm to tasteFollowing blogs and YouTube channels in one readerClustering related coverage by topic instead of source

Featured in

Alternatives to Sprinklz

View all

Reviews (8)

A
Amara Costa

Two months in, no regrets

Picked Sprinklz for the price, stayed for the quality. Where it really wins is nlp clustering groups related stories by topic. Setup was painless and I was productive the same day. No regrets so far.

6/7/2026 13 found this helpful
A
Anders Weber

Quietly excellent

Came to Sprinklz after getting frustrated with what I had before. It handles the boring parts so I can focus on the work that matters. It has shaved real time off my week. Found it works best for clustering related coverage by topic instead of source.

5/29/2026 9 found this helpful
E
Emerson Iyer Verified

Quietly excellent

Sprinklz has quietly become part of my daily flow. Their take on in-browser source code editing is genuinely good. It earns its place in my stack.

3/30/2026 8 found this helpful
Z
Zahra Schneider Verified

Solid daily driver

Three months of Sprinklz later, here is what holds up. What stands out is how little babysitting it needs. It has shaved real time off my week. Worth it for what I get out of it.

5/4/2026 7 found this helpful
J
Javier Weber Verified

Two months in, no regrets

Sprinklz has quietly become part of my daily flow. Where it really wins is configurable recommendation engine. The output quality holds up better than I expected. It fits well for rewriting a hacker news ranking algorithm to taste. Would sign up again without thinking twice.

4/15/2026 6 found this helpful
L
Leon Santos

Worth a look

Tried Sprinklz on a side project first, then rolled it out everywhere. Their take on nlp clustering groups related stories by topic is genuinely good. It fits well for rewriting a hacker news ranking algorithm to taste. Glad I made the switch.

7/5/2026 4 found this helpful
D
Diego Ito

Genuinely impressed

Sprinklz has quietly become part of my daily flow. Got real value out of fully customizable themes and styles. It fits well for rewriting a hacker news ranking algorithm to taste. It earns its place in my stack.

6/7/2026 3 found this helpful
R
Ryota Nowak Verified

Finally something that fits

Three months of Sprinklz later, here is what holds up. The topic similarity and clustering is more useful than I expected. It fits well for following blogs and youtube channels in one reader. Recommending it to people in a similar spot.

4/21/2026