https://nommer.ai/ has a waitlist, planning to ship v1 by end of month.

For the past few months I've been building a Mac Menu Bar app that reduces your claude code token usage costs by ~50%.

This unlocks about 2x more usage within your same Claude Code plan.

Users seem to like it so far.

60 DAUs who together have saved 10.5B tokens so far, representing about $35k in savings.

I'd love for the community to try it out and get your feedback. It's admittedly still a bit rough around the edges so any bug reports are more than welcome too!

Comments URL: https://news.ycombinator.com/item?id=48322017

Points: 2

# Comments: 0

Menu bar app that reduces your Claude Code token costs by ~50% so you get 2x more usage out of your plan.

People seem to like it so far :-)

1. On install we explain what Headroom installs 2. We added an uninstall feature that removes all of this for you 3. On quit of the app, we immediately remove all items that may intervene with normal Claude Code behavior

I did want to call out that headroom is not based on RTK - it includes RTK sure, but headroom cli has a lot more going on under the hood. For more see https://github.com/chopratejas/headroom

https://github.com/gglucass/headroom-desktop (mac app)

https://github.com/chopratejas/headroom (cli)

Did you already try tools that can help to reduce token usage cost so you can get more prompts in within your same plan? Some great ones are

https://github.com/gglucass/headroom-desktop

https://github.com/rtk-ai/rtk

https://github.com/chopratejas/headroom

https://github.com/samuelfaj/distill

Frustrated with Claude Code usage limits I stumbled upon the awesome Headroom project by Tejas Chopra. After jumping through the hoops to install it, I realized the majority of devs don't necessarily use python and aren't comfortable just installing pip packages so I set out to create a Mac App wrapper with Tejas blessing. Here it is!

What Headroom does: It sits between your editor and Claude, compressing what gets sent — trimming redundant context, stripping boilerplate, and compressing documents — without dropping anything semantically meaningful. The result is roughly 2x more usage out of the same Claude plan.

Everything runs locally. Your prompts never touch our servers.

Pricing: Free tier available. Paid plans start at $5/mo and are tiered to match Claude's own plan structure (Pro, Max x5, Max x20). 14-day trial on signup.

Happy to answer questions about how the compression works, or anything else.

Comments URL: https://news.ycombinator.com/item?id=47602591

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47233651

Points: 1

# Comments: 0

What do you base purchase decisions on?

Ultimately it’s imo more about avoiding bad products than it is about noticing whether a product is great. Criticaster collects all professional reviews to get to an average critic score, which will more quickly and more accurately get you to a satisfactory product than any other approach.

In my experience buying a product that gets mostly positive reviews from professional reviewers gets you a far better product than one with mixed reviews. I’m surprised that’s not your experience and that you don’t much trust those scores. What else do you make your purchase decisions on then? Purely specs?

Comments URL: https://news.ycombinator.com/item?id=47170697

Points: 2

# Comments: 10

Comments URL: https://news.ycombinator.com/item?id=47167925

Points: 7

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47060908

Points: 1

# Comments: 0

I often wished something like this would exist for physical products as right now I’m usually stuck doing days of manual research online myself.

Over the past 4 weeks I’ve been building an experiment that applies the same aggregation idea to tech products. It collects professional reviews, extracts and normalizes scores, and produces a single “critic score” per product.

So far the dataset includes ~9,630 reviews across 1339 products. As a small sanity check, I compared the results against two recent purchases of mine, and the “best for most” recommendation matched what I eventually chose after many hours of manual research.

I’m curious what you think about this approach, especially around score normalization, bias between publications, and whether you agree that a single aggregated score is super useful when evaluating products.

Comments URL: https://news.ycombinator.com/item?id=46989556

Points: 4

# Comments: 1

A metacritic like website but for any product.

It analyzes thousands of professional critic reviews to find the best of the best.

I started building this because I adore how metacritic analyzes professional movie/game/tv show reviews and calculates a meta score for each title. In my experience it’s the best way to discover new things to watch or play, and I’ve often wished something like this would exist for when I want to buy a product.

This year, I decided to start building it myself and Criticaster is the result.

For a given product category we collect all professional reviews of a given product, analyze each to assign them a score and then calculate an average critic score.

The goal is to become the most trustworthy source to make product decisions.

Very curious what y’all think!