How I Built an AI-Powered Job Search Radar to Cut Through the Noise

The Job Search Information Overload

Every morning, I used to wake up to a flood of job alerts, LinkedIn notifications, and recruiter emails. Each one claimed to be the 'opportunity of a lifetime.' But after spending 20 minutes scrolling, I felt like I knew a lot but had no clear idea of what actually mattered. Sound familiar? The problem isn't a lack of job openings—it's that there are too many, and they all look the same.

Then I realized: the same techniques I used to curate tech news could work for job hunting. So I built a system, a sort of AI-powered job radar, that filters, verifies, and organizes opportunities so I can focus on what truly fits. Here's how you can do it too.

Step 1: Stop Relying on Random Job Alerts

My first attempt was simple: set up a daily 8 AM task where an AI agent would scrape five job postings and send them to me. It worked—kind of. I'd get a neat summary of new roles while making coffee. But after a few days, I noticed problems. The same job would appear three times under different titles, or an old posting would resurface as 'new.' The AI was doing its job, but the inputs were messy.

The fix? Curate your sources. I had already collected 161 RSS feeds for news, so I applied the same logic to job hunting. I subscribed to company career pages, industry job boards, and niche recruiters. The key is to prioritize primary sources—official company postings—over aggregators that just reshare listings.

Build a Tiered Source Pool

Think of your job sources like a newsroom. You want the most reliable first. Here's my hierarchy:

• Primary sources: Company career pages, official LinkedIn company pages, and government job portals. These give you unvetted, original listings.
• Secondary sources: Industry-specific job boards like Dice, Hacker News 'Who's Hiring,' or niche communities. They often have unique roles not posted elsewhere.
• Tertiary sources: General aggregators like Indeed or Monster. Use them for broad searches, but always cross-check with the original company posting.

This tiered approach cuts down on duplicate and outdated listings. You start to see patterns—which companies are hiring aggressively, what skills are in demand—without the noise.

Step 2: Organize Your Job Feeds Like a Custom Magazine

Once you have dozens of sources, you need a way to manage them. I use an RSS reader called Folo, but any good reader works. The idea is to organize your sources into categories: 'Tech Startups,' 'Remote Jobs,' 'Finance Roles,' whatever fits your search. This way, you can skim your 'magazine' quickly, focusing on the sections that matter most today.

For example, I have a folder for 'AI Companies' and another for 'Hardware Firms.' When I open Folo, I see a clean, magazine-like layout. It's not just about reading; it's about controlling what enters your field of vision. You're not at the mercy of an algorithm that shows you the same five jobs because you clicked once.

Step 3: Teach Your AI to Act Like a Managing Editor

The real game-changer was connecting my RSS reader to an AI agent. Folo has a CLI tool, so I could program the agent to read my unread job postings and apply 'editorial judgment.' I gave it these rules:

• Merge duplicates: If two sources list the same job, output it once.
• Cross-verify details: Compare salary, location, and requirements across sources to give me the most complete picture.
• Prioritize primary sources: If a posting exists on the company's site, use that as the base and note 'verified.'

This turned my AI from a simple scraper into a managing editor. It now sends me a daily digest of five genuinely distinct, relevant jobs. No more duplicate listings, no more old posts resurfacing.

Step 4: Customize Your AI's Preferences—It's Okay to Nag

Even with clean sources, the AI didn't know what I valued. At first, it kept showing me senior-level roles when I was looking for mid-level positions. So I told it directly: 'Show me more product management roles with 2-5 years experience, and less data science.' The AI stored this in a memory file and adjusted future digests. It's like training a new assistant—you have to correct it a few times before it gets it right. But once it does, the results are spot-on.

Step 5: Create Your Own 'Cyber Job Newspaper'

Text summaries are fine, but I wanted a visual dashboard. Since the AI can code, I asked it to generate an HTML page with my daily job digest. It creates a clean, card-based layout with sections for 'Title,' 'Core Facts,' and 'Why It Matters.' Each card has a button to the original posting. It's like a personalized job newspaper that updates every morning.

I also used this to track a specific role type over time—say, 'foldable phone hardware engineer'—and the AI built a dynamic tracker with a timeline, source credibility tags, and a keyword frequency chart. It showed me how often certain skills appeared and which companies were consistently hiring for that role. That's powerful insight for tailoring your applications.

Why This Matters for Your Job Search

In 2025, the word 'slop' was chosen as Word of the Year, referring to low-quality AI-generated content. Job listings are no exception. Many are auto-generated, vague, or even fake. By building your own curated feed, you're building a dam against the flood of digital garbage. The AI can collect and deduplicate, but you keep control of your judgment. You decide what's worth your time.

Final Thoughts: Build Your Own Dam

You don't need to be a tech wizard to set this up. Start with a handful of trusted sources, use an RSS reader to organize them, and let an AI agent do the heavy lifting of filtering and summarizing. The goal isn't to see every job—it's to see the right jobs. When you can trust your feed, the anxiety of missing out fades. You'll know what to read, what to skip, and where to focus your energy.

So, take back control. Build your own job search radar. It might just be the most productive thing you do all week.