About HiveCode
HiveCode is a small independent studio building focused web tools. This page is the longer version — who we are, how we work, and what you can expect from anything we publish.
How we got started
HiveCode began as a small set of personal utilities — duration converters, timecode helpers, micro-formatters — that we kept writing one-off versions of for every project. At some point it was easier to publish them at clean URLs than to keep dropping the same snippet into different repos. That is the whole origin story; we got tired of opening the calculator.
The studio name comes from the metaphor we use to talk about how we build: a hive of small, finished cells. Each tool is its own cell — separate URL, separate codebase, no shared signup, no shared backend. The hive holds them together but each one stands on its own.
Who works on this
HiveCode is a few colleagues and friends working out of a small town in northwest Colombia. We have worked together long enough that the studio reads more like a workshop than a company, and that is how we like it.
The team is deliberately small. Small enough that the person who designs a tool is the same person who deploys it, and the same person who answers your email if you find a bug. We do not disclose exact size because it varies between contractors and full-time, but the upper bound is comfortably in single digits.
We work across English and Spanish. Spanish is our day-to-day; English is for the rest of the internet. Both languages are first-class on every tool we ship — translations are written by humans on the team, not auto-generated from a single source. If a phrase reads awkwardly in either one, that is a bug we want to hear about.
The shape of a HiveCode tool
Every tool starts the same way: we hit a recurring task in our own work and want one less tab open. From there the rule is to keep the scope small — one input, one or two outputs, no settings panel, no account, no growth funnel. If the brief grows, we split it into a second tool.
Once built, the work does not stop. We watch for accessibility issues, slow loads, locale edge cases (yes, decimal commas), and we ship fixes when they come up. There is no roadmap of artificial features; the only updates are the ones that make the existing tool better.
We do not promise tools we have not started. The current published catalog is what we have. When something new is ready, it shows up at hivecode.cloud and gets its own subdomain. No teasers.
What we build with
All HiveCode tools are statically generated with Astro, served from a small server through nginx, and hosted on infrastructure we control. There is no client-side framework runtime — the JavaScript on the page is whatever the tool itself needs, and nothing else. A typical page is under 50 KB on the wire, including fonts.
We use plain CSS, system-stack-leaning typography (Hanken Grotesk + Geist Mono), and OKLCH color tokens to keep palettes perceptually consistent. The code lives in public GitHub repositories under the HiveCode-X organisation when it makes sense to publish.
What you can expect
Every HiveCode tool will: run in your browser without an account, work after first load even if your network drops, ship with bilingual UI from day one, expose a privacy page that says exactly what is happening, and keep advertising — when used — to a single placement that does not interrupt the tool.
When we change any of this, we update the relevant tool’s privacy page first. The date there is authoritative.