Hi HN,

Firs HN post here. I’ve been reading and following HN for years, so I’m really curious to hear feedback from you :)

API Unit started out of frustration, not as a startup idea.

In several projects, our APIs were no longer simple request-response endpoints. A single business flow meant multiple dependent calls, shared state, conditional logic, retries, and different outcomes depending on responses.

At first, we handled this with Postman collections. That worked… for a while. Then collections grew, scripts multiplied, monitors became detached from the actual test logic, and it became hard to answer simple questions like: - Which flows are failing consistently? - What exactly ran last night? - Which part of the chain broke and why?

We realized the problem was not “sending requests”. It was orchestrating API test flows over time.

So I built API Unit around a different assumption API testing is closer to workflow orchestration than to clicking requests. With API Unit you model API tests as flows, group them into suites, and schedule them to run automatically. When something fails, you get an alert on mail, a clear execution history showing how the flow progressed and where it broke.

It’s REST-only and intentionally not an API client replacement, but you a page dedicated to simple requests as well. The tool is meant for teams that already know their APIs and want reliable, repeatable testing and monitoring of real-world flows. The product is live and still evolving. I’m sharing it here to get honest feedback from people who have dealt with growing API test complexity, especially around chaining, scheduling, and long-term visibility.

You can create account for free and test everything for 3 days, no card required. Also if you wanna use API Unit, send me a mail at contact@apiunit.io with your account email. I have a gift for you. :)

Website: https://apiunit.io

Happy to answer questions or hear why this wouldn’t work for your setup.

Elon Musk is getting serious about orbital data centers

San Francisco is seeing a jobless AI boom

Team unveils simpler, faster way to make vaccines

"Slow Tuesday Night" by R. A. Lafferty (2015)

Django Reinhardt

O Que é Desenvolvimento Pessoal E Profissional de Verdade

It's 2026, Just Use Postgres

Epstein Graph Visualization

Recursive Self-Improvement (Part I)

Live agent face-off in CivBench: Claude Opus 4.6 vs. GPT-5.2

Infrastructure configuration can swing coding evals by several % points

Preserving the Open Web: Inside the New Wayback Machine Plugin for WordPress

I'm in the Epstein Files

Django: Profile Memory Usage with Memray

Instacloud as infinite cloud storage using Instagram as remote disk

Calling Lean Functions as Python Functions

Expertise is a Relic; They want Drones

Fli4l

Why AI-Generated Code Will Hurt Both Customers and Companies

Voxtral.c Voxtral Realtime 4B model inference as a C library

Aqua2Terra: Bringing subsea cable architecture to terrestrial fiber networks

Ottermon.ai – Effortless Observability Deployed in Seconds

Learning to Reason in 13 Parameters

The Logic of Surveillance (2013)

Claude Code Is the Inflection Point

Experimental: Add xx-zones protocol for area-limited window positioning

Show HN: API Unit manage and schedule real API test flows, not just requests

Elon Musk – "In 36 months, the cheapest place to put AI will be space" [video]

Bringing Engineering-as-Code to the Sphinx Framework

BalatroBench – Benchmarking LLMs' Strategic Performance Through Games