cd /news/ai-agents/show-hn-railcode-secure-cloud-for-in… · home › topics › ai-agents › article
[ARTICLE · art-147193] src=railcode.dev ↗ pub= topic=ai-agents verified=true sentiment=↑ positive

Show HN: Railcode – Secure cloud for internal software

Yakko launched Railcode, a secure internal deployment platform for company software, on Hacker News' Show HN. Railcode places company authentication in front of every deployed project, replaces shared API keys with permissioned data connectors, and adds centralized audit logs and role-based access control, addressing the sprawl that occurs when employees build tools with Claude Code or Codex and deploy them to services like Vercel. The platform supports static frontends and serverless JavaScript functions, including Tanstack, Hono, Vite, React, and Next.js via OpenNext, and ships a skill and CLI for agents to manage configuration.

read2 min views3 publishedOct 7, 2026

Hey HN, Yakko here from Railcode.

Railcode is like an internal Vercel for your company.

We built it after speaking to dozens of engineering leaders at various companies where employees were given access to Claude Code or Codex and started to build little tools (apps, agents) locally, but then struggled to share it with coworkers securely.

Typical scenarios are:

  1. Employees can't share what they've built because of internal policies

  2. Agents find the path of least resistance and deploy things on e.g. Vercel, leading to a sprawl of software that's a nightmare to manage (disjointed auth or no auth, API keys handled insecurely, no centralized management, etc.)

  3. Companies set up templates to ensure some standardization but still require engineers in the loop to review and deploy code

Neither of these options felt great, so we built Railcode to be the place to deploy and manage internal software inside a company, ensuring projects are deployed securely, follow the same patterns, and can be easily managed from one place.

The goal is to unblock users to deploy and share what they've built while removing the burden on Engineering/DevOps to review and manage it all.

How we do this:

  • Company auth in front of everything that's deployed

  • Permissioned connectors to access data instead of shared API keys

  • Infrastructure building blocks that make it easy to store data or use an LLM but allows admins to oversee everything (via a simple SDK)

  • Centralized audit logs

  • Unified design system (optional)

  • Role-based access control with granular permissions that can be inherited by the projects themselves

The platform is built to be really low friction for agents and to not require folks changing their workflows. We support static frontends and serverless JS functions for the backend, meaning Tanstack, Hono, Vite, React, etc will all work. Even Next.js with OpenNext though that has some quirks :D

Projects run locally and everything Railcode-specific comes from an optional SDK. We ship a skill and a CLI and that's all you need to build and deploy projects or have agents manage configuration.

Keen to hear feedback!

Comments URL: [https://news.ycombinator.com/item?id=49999322](https://news.ycombinator.com/item?id=49999322)

Points: 1

── more in #ai-agents 4 stories · sorted by recency
── more on @railcode 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/show-hn-railcode-sec…] indexed:0 read:2min 2026-10-07 · —