Launching today
Lattice Grid
JS Data Grid With Charts, Kanban, Gantt, Stats and Maps
2 followers
JS Data Grid With Charts, Kanban, Gantt, Stats and Maps
2 followers
Lattice Grid is a JavaScript data grid that is also the rest of your data UI. Show your rows as a grid, chart, Kanban, Gantt, calendar, map or form, all reading the same data, and let the data router split one feed so a single query fills every view on a screen. Sorting, filtering, grouping, pivots, tree data and validated editing are built in, and it scrolls a million rows smoothly. Use plain JavaScript from one script tag, or React, Angular, Vue, Svelte or HTMX.
Free Options
Launch Team
CloudflareYour next app belongs on Cloudflare Workers. Deploy for free
Promoted
Hi Product Hunt 👋 I’m Tony, and I built Lattice Grid.
It started with our own product. We were building an telecoms SAAS and, like most teams, we assembled the UI from parts: one library for the data grid, another for charts, another for the Gantt, another for the Kanban board, a map library, a form library. Each had its own data model, its own idea of a date, its own validation and its own theming.
The result was glue. Code to keep a chart in sync with the grid’s filter. Code to make the form accept exactly what the cell accepted. Code to move a selection from the map to the table and back. Every upgrade of one part broke a seam somewhere else, and most of our bugs lived in the glue, not in our product.
So we built our own set of tools on one data layer:
One dataset, many views. The grid, charts, Kanban, Gantt, scheduler, calendar, maps and forms all read the same rows. Filter the grid and the chart follows; edit a card and the row changes.
Statistics where the data is. Summary stats, percentiles and confidence intervals on any column; correlation and regression between columns; outlier flags you can style in the cell; forecasts and seasonal breakdowns; histograms, box plots and Pareto charts; clustering; and every figure compared against the full population, not just the rows on screen. No export to a notebook first.
A dashboard designer for your users. Embed a report builder in your app: you declare the sources, how they relate and the guardrails, your users build their own dashboards, and you get each one back as plain JSON to save.
Tools for the people using it. Full screen on a button. Draw over the grid with a pen, arrow, box or highlighter while you talk it through. Presentation mode scales the screen to a room and steps through saved views like slides, still live. And redact a column before a screen share: the row count and the sort stay, only the values go.
The data router. One feed in, every view filled. It splits a single query, socket or stream by type and sends each part to its own grid or chart, so one round trip hydrates a whole screen.
Multiple data sources out of the box. Plain arrays, CSV files, Parquet queried in the browser through DuckDB, live socket streams, or your own server doing the paging.
Fast. It scrolls a million rows smoothly, and the local-file demo opens 10 million.
It’s plain JavaScript from one script tag, with adapters for React, Angular, Vue, Svelte and HTMX. It’s free to use on localhost; you buy a licence per domain when you ship.
Try out a demo here: https://toclocoinc.github.io/lattice-grid-demo-house-prices-umd/
What would help us most: try it on your own data and tell me where you still had to write glue. That’s the gap we want to close next.