@ycombinator: Radio Frequency test equipment, finally built like it's 2026. Zero driver installs, PoE and USB-C, automatic device dis…
Summary
Nine Fives launches driverless, networked RF test equipment with modern interfaces like PoE and USB-C, aiming to streamline production testing for radio device manufacturers.
View Cached Full Text
Cached at: 05/22/26, 09:45 AM
Radio Frequency test equipment, finally built like it’s 2026. Zero driver installs, PoE and USB-C, automatic device discovery and networking, and a SKILL.md to write your test software.
Smash your test development schedule and ship to production faster with @NineFivesRF.
https://ycombinator.com/launches/QQk-nine-fives-simple-networked-driverless-rf-test-equipment…
Launch YC: Nine Fives - Simple, Networked, Driverless RF Test Equipment | Y Combinator
Source: https://www.ycombinator.com/launches/QQk-nine-fives-simple-networked-driverless-rf-test-equipment Test hardware for companies that can’t wait for their test equipment to catch up.
TL;DR: Nine Fives makes test equipment for manufacturers of radio devices in cellphones, drones, or satellites.
https://www.youtube.com/watch?v=7vtMt6-0SvY
**Problem:**You are a hardware engineer holding up a billion dollar satellite constellation or slipping a cellphone release date because the legacy test equipment in your production line was designed around the January 1997 revision of the USB Human Interface Device specification and the firmware hasn’t been updated since then.
Your AI agent is blissfully unaware of how this thing works because the “documentation” is hidden in a PDF on a password protected FTP server. After the legacy vendor left your email on read for a week they gave you access and sent you aVB.NETscript that they think should work. You ran it and now all 8 of that vendor’s devices in your rack have tried to enumerate as “COM11” in device manager. Now you have 8 broken pieces of equipment instead of just 1!
**Solution:**We’re Noah Levy and Andrew Kurtz and we are building Nine Fives. Together we have spent a decade at SpaceX designing radio frequency hardware and test systems for multi-million dollar products. Our equipment is driverless, touchscreen enabled, and uses a single, machine-readable, API with both REST and SCPI endpoints.
Every module we build works as a standalone product, but if you use our rack-mountable data and power backplane than you can take advantage of our NineVue web-app for generating easy-to-maintain, machine-readable rack configuration and ourskill.mdso that your AI agent can finally understand your test system and write your test software.
Get off the critical path and reduce your production readiness timeline from months to minutes.
Similar Articles
@ycombinator: http://Prototyping.io is building autonomous manufacturing systems to turn CAD designs into high-quality mechanical par…
Prototyping.io, a Y Combinator-backed startup, launched an autonomous manufacturing platform that uses AI agents to automate the process from CAD design to finished mechanical parts, claiming $400k monthly revenue and significant iteration cycle savings for enterprise customers.
Launch HN: Simantic (YC F26) – Firmware Simulation for AI Agents
Simantic (YC F26) Launch HN post announcing a firmware simulation layer that lets engineers and AI agents test and validate firmware on virtual devboards without physical hardware, positioned as a faster alternative to Renode and QEMU.
@AnjneyMidha: very cool
Etched announced it is coming out of stealth after successful A0 tapeout, with $1B+ customer contracts and $800M raised, claiming SOTA inference performance on its first racks shipping this summer.
Building a mini Homelab that fits in my carry-on
Jeff Geerling built a portable homelab in a carry-on suitcase using Ubiquiti UniFi networking gear for a demo at VCF Midwest, featuring high-speed networking, 5G backup, and travel-ready design.
RFBench: A Simulator-Grounded Multimodal Benchmark for Passive RF Layout Design and Understanding
RFBench introduces a simulator-grounded multimodal benchmark of 10,000 passive RF layouts with seven evaluation tasks, revealing systematic weaknesses in multimodal language models at generating structurally valid layouts and predicting electromagnetic responses, while RF-specific fine-tuning yields substantial gains.