DEV Community: Guy Ruvio

How I Built an AI Job Board That Auto-Updates from 164+ Companies

Guy Ruvio — Wed, 15 Apr 2026 07:36:11 +0000

The Problem

If you're looking for an AI engineering role in 2026, you're probably checking 10-20 individual career pages. Anthropic's Greenhouse page, OpenAI's careers site, DeepMind's jobs board, Cohere's Lever page...

General job boards like LinkedIn and Indeed have AI roles, but the filtering is terrible. Search "AI engineer" and you get results for "AI-powered customer service" and "engineer at an AI startup doing nothing related to AI."

I wanted a single page with every AI/ML/LLM role from every major AI company, updated automatically. So I built LLMHire.

Architecture

Stack: Next.js 14 (App Router) + Supabase (PostgreSQL) + Vercel (hosting + crons)

The core idea is simple: most tech companies use one of three ATS platforms — Greenhouse, Ashby, or Lever. Each exposes a public API for their job listings. I built adapters for each.

The Fetcher Pattern

Each ATS has a fetcher module:

Greenhouse: GET https://boards-api.greenhouse.io/v1/boards/{company}/jobs
Ashby: POST https://jobs.ashbyhq.com/api/non-auth-job-posting
Lever: GET https://api.lever.co/v0/postings/{company}

Each fetcher normalizes the response into a common schema:

title, company, location, department
remote/hybrid/onsite classification
role type (ML Engineer, LLM Engineer, Research Scientist, etc.)
posting date, description, apply URL

Auto-Categorization

Job title → role type mapping uses pattern matching:

Contains "LLM" or "NLP" or "language model" → LLM Engineer
Contains "ML" or "machine learning" → ML Engineer
Contains "research" + "scientist" → AI Research
Contains "infrastructure" or "platform" → AI Infrastructure
Contains "prompt" → Prompt Engineer

This gets ~90% accuracy. Edge cases get manually reviewed.

Cron Jobs

Vercel cron jobs run every 4 hours (6x/day):

Fetch all jobs from all companies
Deduplicate against existing listings (by company + title + location hash)
Mark stale listings (>30 days without refresh) for review
Update search index

The Numbers

130+ companies tracked
670+ active positions
~50-100 new/updated listings per day
Average fetch cycle: ~45 seconds for all companies

What the Data Shows

After tracking the AI job market for a few weeks:

"LLM Engineer" is a real role now — distinct from ML Engineer, with different skill requirements
Remote is declining — major labs want hybrid/onsite
AI Infrastructure is booming — fastest growing subcategory
Prompt Engineering peaked — being absorbed into broader roles

What's Next

Workable integration (another ATS with significant AI company coverage)
Salary data aggregation
Weekly AI jobs newsletter
Blog with hiring market analysis

Try It

LLMHire is completely free. No accounts, no fees, no signup required to browse.

If you have feedback or companies I should add, drop a comment.

Anthropic Accidentally Published 513K Lines of Claude Code Source on npm — What Developers Need to Know

Guy Ruvio — Thu, 09 Apr 2026 23:06:37 +0000

On March 31, 2026, a 59.8 MB JavaScript source map shipped inside the npm package @anthropic-ai/claude-code version 2.1.88. It contained 513,000 lines of unobfuscated TypeScript across 1,906 files — Anthropic's full Claude Code agent architecture, published by accident.

What Happened

A Bun packaging error combined with a missing .npmignore file caused the source map to be included in the published npm package. Security researcher Chaofan Shou discovered and posted about it on X. Within hours, the source was mirrored to GitHub and forked tens of thousands of times.

The CVEs

Two CVEs were assigned:

CVE-2025-59536 — Remote code execution via malicious repository configs
CVE-2026-21852 — API key exfiltration through hooks and MCP (Model Context Protocol) servers

The leak exposed exact orchestration logic, making these exploits trivially reproducible by attackers who studied the source.

The Same-Day Supply Chain Attack

In a coincidence that made an already bad day worse, the axios npm package was trojaned with a Remote Access Trojan (RAT) between 00:21 and 03:29 UTC on March 31 — hours before the Claude Code leak became public. If your CI/CD ran npm install during that window, you may have pulled a compromised version of axios alongside Claude Code.

What You Should Do Now

Audit your npm install logs for March 31 — check what versions were pulled
Check for axios RAT indicators — the trojaned version was active for ~3 hours
Scan repos used with Claude Code — hooks and MCP server configurations are the primary attack surface
Update to the latest Claude Code version — the source map has been removed

Video Breakdown

I put together a detailed breakdown covering the full timeline, what was inside the leaked source, and specific steps for developers:

Sources

The Hacker News — Claude Code leak via npm (April 2026)
Zscaler ThreatLabz analysis
Bloomberg, April 1, 2026
Axios RAT supply chain attack report

Originally published on EndOfCoding.com. Follow @endofcoding for more AI security analysis.