These episodes of #thePOZcast, live from Transform 2026 in Las Vegas, are proudly brought to you by our friends at PIN. AI recruiting tools that automate candidate sourcing, screening, and scheduling across 850M+ profiles. Built for recruiters, agencies, and hiring teams.

Learn more and check out a demo:  https://www.pin.com/book-a-demo?via=adam-posner

Thanks for listening, and please follow us on Insta @NHPTalent and www.youtube.com/thePOZcast (http://www.youtube.com/thePOZcast)

For all episodes, please check out www.thePOZcast.com (http://www.thepozcast.com/)

TAKEAWAYS: 

1. Same Tools, Same Results — You Have to Rebuild the Engine

The insight at the heart of Pin: giving AI the same Boolean search infrastructure that human recruiters use produces the same mediocre results, just faster. The only way to get genuinely better outcomes is to rebuild the search engine itself so that AI can operate on a fundamentally different foundation. That's what Pin did — and why the results look different.

2. The Best Candidate Should Be First, Not on Page Seven

The clearest signal that a recruiting search tool is working: the most qualified candidate for a role appears at the top of results, not buried deep in a list that requires manual excavation. For recruiters who've spent years digging through pages of search results, seeing the right person in slot one is a genuinely disorienting experience — in the best way.

3. Natural Language Filtering Closes the Gap Between Search and Judgment

Standard filtering tools handle objective criteria — location, tenure, title. Pin's natural language feature handles the subjective judgment calls that used to require hours of resume scanning: the specific details that determine whether a candidate is actually worth a call. Resolving those questions in two questions or fewer is a meaningful time return for high-volume recruiters.

4. Pattern Recognition Learns Even Without Feedback — But Feedback Makes It Faster

Pin's algorithm doesn't require explicit feedback to improve — it reads behavioral patterns in what recruiters accept and reject and adjusts accordingly. But providing reasons for rejections accelerates the learning dramatically. The system is watching, learning, and tuning, whether or not you tell it why.

5. The Curveball Candidate Is a Feature, Not a Bug

Periodically surfacing a candidate who sits just outside the current search parameters isn't an error — it's deliberate calibration. When a recruiter declines that candidate, Pin learns where the line actually lies, resulting in increasingly precise results over time. The tool is always running a low-stakes experiment to get better.

6. A Visual Pipeline Changes How You Manage a Search

Pin's upcoming Kanban board — drag-and-drop stages from interested through offer made — addresses one of the most persistent frustrations in recruiting: knowing at a glance where every candidate stands without digging through notes or spreadsheets. Pipeline visibility is a workflow problem as much as a sourcing one.

7. MCP + Claude Desktop = Autonomous Sourcing

The MCP Server integration is the most forward-looking announcement in this episode: the ability for Claude Desktop to run Pin autonomously, without manual recruiter input, using Claude's broad knowledge base to execute searches and surface candidates. For business development and high-volume sourcing, this is autopilot for the top of the funnel.

8. The Second Company Is Easier Because the Team Already Knows How to Build Together

Steven's team story is a blueprint for founder-led companies: seven people from his first venture joined him at Pin, bringing a shared language, shared trust, and a shared understanding of what works and what doesn't. The result is what Steven calls "life on easy mode" — not because the work is easier, but because the team already knows how to do it together.

9. Always Give Feedback to Your AI Tools

Every rejection is a data point. Every accept is a signal. The recruiters getting the best results from AI-powered search tools are the ones who treat the interface as a two-way conversation — providing context, reasons, and reactions that train the system toward increasingly precise output. Passive use gets passive results.

 

CHAPTERS:

00:00 – Day 9: The Return of Steven Lu Adam, on day 9 of 10 at Transform, welcomes back Steven Lu — a returning guest and the founder of Pin, the recruiting AI tool Adam uses every day.

02:00 – Why Giving AI Boolean Tools Gets You Boolean Results The core problem Pin was built to solve: if you give AI the same search tools as a human recruiter, you get the same results. Pin rebuilt the search engine itself so AI could actually deliver better outcomes.

04:30 – The Aha Moment: Best Candidate, Slot Number One What clients experience when they switch to Pin: the best candidate for the role appears first — not buried on page seven.

06:30 – Natural Language Q...