Show HN: Matrices – Explore, visualize, and share large datasets

[calvinfo]

Hey HN, I'm excited to share a new side project I've been working on.

The product is called Matrices. You can check it out here: https://matrices.com/.

With Matrices, you can explore, visualize, and share large (100k rows) datasets–all without code. Filter data down to just what you want, visualize it with built-in charts, and share your results with one click.

You can use it today (no login or waitlist or anything). Just copy and paste your data from a google sheet or CSV file.

It's hard to describe the feeling of "gliding over data" you get with Matrices, so I'd rather _show_ you how it works instead. This 75s video will give you a sense of how it works: https://www.youtube.com/watch?v=Rrh9_I3Ux8E.

Data is stored locally in your browser until you publish it, though small sample does go to the OpenAI APIs for AI-assisted features.

I started building Matrices because I wanted a tool that made it easy to explore new datasets. When I'm first trying to dig into data, I'll have one question... that leads to another... that will invariably lead to five more questions. It's sort of a fractal process, and I couldn't find many good options that were fast, responsive, and visual.

I figured this crowd would be interested in tech stack as well, it's using arquero [1] bindings over apache arrow for in-memory analytics, and visx [2] for visualizations. I'd like to add duckdb-wasm support at some point to open up a wider set of databases. Data is serialized as parquet to save a bit on bandwidth + storage.

Give it a spin, and let me know what you think. This is my first 'serious frontend project' so I appreciate any and all feedback and bug reports. Feel free to comment here (I'll be around most of the day), or shoot me a note: [e-mail]

[1]: https://uwdata.github.io/arquero/ [2]: https://airbnb.io/visx/

[willium]

This is cool! (how’d you get that domain?) Kind of like gist for data exploration.

Super refreshing to be able jump right in — it feels like every tool is behind a “talk to sales” CTA these days.

What are you using for the charts? 100k points on a scatter is pretty impressive with SVG.

[calvinfo]

Thanks!

Domain was an opportune buy off Flippa. It wasn't really being used for anything and the seller wanted to get rid of it.

> What are you using for the charts? 100k points on a scatter is pretty impressive with SVG.

Visx/SVG right now. You're right that 100k points gets slow on a scatter plot, for those charts in particular some sampling occurs. At some point I'd like to investigate other options, but this works for the MVP.

[bake]

This is impressive -- Google sheets stalls out for me all the time with data sets that are pretty small, all things considered. What was your trick for making it so much faster inside the same browser? Anything design decision that stands out?

[calvinfo]

A couple of things that really help...

Matrices isn't editable (sort of by design) and works best for columns that are all of the same type. It then uses arrow for fast in-memory analytics.

Google sheets has to serve a much broader array of use cases, so I think they can only do so much to improve the performance. It can't always rely on having consistent rows and columns.

[cdr_fyi]

This is wicked fast - one of the biggest problems with Google sheets is dealing with large datasets