Sourcing From HN's 'Who Wants to Be Hired': A Pool LinkedIn Can't Index

If you source engineers on LinkedIn, from November 2026 you'll be ranking the same crowded pool with the same AI as every other Hiring Assistant customer. Meanwhile a pool of free, structured, self-declared candidates on Hacker News sits mostly unread.

A dataset published on dev.to ran 13 months of HN hiring threads through an LLM. Company posts fell by a third over that stretch, and the partial October 2026 count for "Who wants to be hired?" logged 566 candidates against 221 hiring companies. That works out to 2.56 opt-in candidates for every employer, each one asked to post an email address.

What the HN hiring dataset actually shows

Company posts in HN's "Who is hiring?" fell about 33% between Q4 2025 and Q3 2026. Over a shorter window, April to September 2026, candidate posts in "Who wants to be hired?" rose 44%. The partial October 2026 count stands at 566 candidates and 221 hiring companies.

The dataset was published on dev.to by Arthur (user arthur_taken). It covers 13 "Who is hiring?" threads (October 2025 to October 2026, 9,194 roles from 3,909 company posts) and 7 "Who wants to be hired?" threads (April to October 2026, 3,586 candidates). An LLM extracted every field. A few caveats before you quote it:

• October is a partial month. The 566 and 221 come from the table, not the headline.
• The headline uses a different series. Its TL;DR says candidate posts went from 384 in April to 553 in September, and the ratio went from 1.2 to 2.2 candidates per hiring company.
• "~240 companies a month" is a Q3 2026 average, not a single month.
• The author sells the pipelines as Apify Actors. That doesn't make the numbers wrong, but the analysis comes with a product attached.

2.56

The ratio more than doubled between April and October, and pay stayed flat. Supply is outrunning demand in this channel. If you have a real role, it's a buyer's market.

Why one month of HN rivals LinkedIn's open-to-work headlines

In Refolk's index, only 198 of the 531,827 people in the US with a current Software Engineer or Senior Software Engineer title write "open to work" in their headline. One month of HN's thread produces roughly 136 to 198 US-based opt-in candidates. The two pools are about the same size.

The HN estimate is derived: take 566 posts and apply the dataset's 24 to 35% US share. The thread also includes non-SWE candidates, so treat this as an order-of-magnitude comparison, not an exact match. Even so, the top of the HN range lands exactly on the index count.

1 in 2,686

Why the LinkedIn number is so small

The two platforms record intent differently. On LinkedIn, "open to work" is usually a private setting only recruiters can see, or a photo frame. Very few people write it as headline text. The biggest employers among the 198 are Amazon/AWS (6 combined), Google (2) and Oracle (2), so the headline signal is spread thin and doesn't cluster at the big brands.

Germany shows the same thing at a smaller scale. Refolk's index has 11 Software Engineers or Senior Software Engineers there with "open to work" in their headline, mostly at small firms such as Workato, MODIFI and goodBytz. The US has 18x as many, but it's a tiny pool in both countries.

HN candidates don't get a private option. To be in the thread at all, they state their intent in public, in a fixed format, next to an email address. That's why a single monthly post produces roughly as many declared candidates as the whole US headline pool. If you're asking where to find software engineers open to work, a public declaration is a stronger signal than an inferred one.

The thread is a senior market, and that favors founders

"Who wants to be hired?" is mostly senior candidates. Between 57 and 63% describe themselves as senior or above, with a median of 10 years of experience when they state it, while only 35 to 46% of the roles posted are senior and up.

The gap runs the other way at the junior end. Junior roles make up only 2 to 5% of postings, so juniors posting in the thread are mostly wasting their time. Senior candidates face the opposite problem: there are more of them than there are senior seats. That mismatch is why founders hiring a senior generalist face the least competition here.

Employers already hire from this community:

• HN user "sahilly" wrote in May 2026: "We've hired three developers from HN and now we're looking for a QA Analyst."
• Sahil Lavingia (a different person) posted in October 2026: "We're hiring software engineers, product designers, and design engineers at the IRS!"

If you need someone with a decade of experience who has already said in public that they're looking, start here.

Remote-only is the filter that decides who replies

Remote work is the main dividing line in the thread. About half of candidates want remote-only, but only 33 to 42% of roles offer it, a gap of roughly 10 to 15 points.

Pay doesn't separate offers much here: the median US software-engineering salary stayed in a $177k to $195k band all year. Location policy does. Remote budget is scarce in this channel, so a company that has it gets an unusually big share of the strongest candidates. How to use that depends on your policy:

1. Remote-first companies: lead with it. Put "remote" in your first line, because half the candidates won't consider anything else.
2. Hybrid or onsite companies: filter for the 38 to 45% who say they're willing to relocate. Don't email remote-only candidates. You'll waste their time and spend reputation in a community where it travels fast.
3. US-only roles: filter the location field first. Only 24 to 35% of candidates are in the US.

How to source engineers from Hacker News, step by step

To source from HN, run the thread through a search tool, filter on the structured fields, verify each candidate against their linked work, and send one personal email per person about a real role.

The October 2026 post asks candidates to use a fixed format, and that format is what makes the thread searchable:

• Location
• Remote
• Willing to relocate
• Technologies
• Résumé/CV
• Email

Here's the workflow:

1. Use the tools the thread links itself. The October post points readers to nthesis.ai/public/hn-wants-to-be-hired and wantstobehired.com. hnhiring.com is another option. Scrolling 566 comments by hand doesn't scale.
2. Filter on Remote and Willing to relocate first. These two fields eliminate the most mismatches.
3. Filter on Technologies next. Be literal. "Rust" in the stack field beats "systems" in a bio.
4. Open every link. Most posts link a GitHub, a personal domain or a CV. That's the actual evidence.
5. Read the username's history. A few minutes of comment history shows how someone reasons, writes, and handles disagreement.
6. Write one email per person. Name the role, the comp band if you can, and the one thing in their post that made you write.

Step 5 is the signal HN gives you for free and LinkedIn's ranking can't index.

How to verify an HN candidate before you email

You can verify an HN candidate with three free signals that sit outside LinkedIn: a public post history, karma, and a link to real work such as a GitHub profile, a personal domain or a CV. Together they answer "is this person real?" without depending on a platform's verification badge.

That question is a top problem in sourcing: 39% of recruiters name not knowing whether an applicant, or their skills, is real as a top challenge. LinkedIn's answer is verification. It has 115 million verified members, and Hiring Assistant 2 brings those signals together. That works, but only inside LinkedIn's walls, and only for people who went through LinkedIn's checks.

HN's signals are independent of LinkedIn. A username with years of technical comments is hard to fake cheaply. A GitHub link with commit history can be checked. A personal domain usually leads to talks, posts or projects. The slow part is joining them up: matching an HN handle to a GitHub account to a current employer to a work email that isn't a throwaway address. That's the step Refolk handles. You describe the person in plain English and get back a ranked shortlist matched across GitHub, LinkedIn and the open web, so the HN post becomes a lead you can check.

When verification fails, it usually fails in one of three ways:

• A brand-new account with no comment history and a generic CV link.
• A GitHub link to a profile with no original commits, only forks.
• A stack list that doesn't match anything the person has written or shipped.

None of these alone rules someone out. Two together are a reason to move on.

The thread's norms restrict posting, not contacting

HN's rules stop agencies and recruiters from posting in "Who wants to be hired?". They don't stop anyone with a real role from emailing candidates. The October post says it plainly: "Agencies, recruiters, job boards, and so on, are off topic here. Readers: please only email these addresses to discuss work opportunities."

So the rule is about where you post, not who you contact. In-house recruiters are welcome on the hiring side: Seema, an in-house tech recruiter at Ojin in Berlin, posts in "Who is hiring?" herself. The rule against recruiters covers the candidate thread, and the email rule covers everyone.

The rule is simple: one real role, one real person, one real email. Anything automated breaks it.

Here's what actually breaks the norms:

• Bulk or automated outreach. Scraping all 566 addresses into a sequence is exactly what the rules prohibit.
• Emails without a role. "Let's connect" or "I'd love to add you to my network" is not a work opportunity.
• Resume harvesting. Collecting CVs for a database you'll pitch to clients later.
• Replying in the thread. Comment replies pitching jobs are off topic. Use email.

The rules apply to each email, and HN reputation travels fast. The safe volume is however many people you can write to personally, not 566.

Where LinkedIn Hiring Assistant 2 fits

Hiring Assistant 2 makes LinkedIn's pool faster to search. It doesn't make it a different pool, which is why sourcing developers outside LinkedIn is worth more now, not less. Over 20,000 companies use LinkedIn's agentic hiring products, and HA2 rolls out automatically from November 2026, at no extra cost, to every existing Hiring Assistant customer working in English.

LinkedIn says v1 users review 83% fewer profiles before finding a qualified match, and the agentic hiring business has passed $450M in annualized revenue. Those numbers will push adoption, which means more recruiters working from the same model's ranking of the same pool.

Ranking doesn't help when everyone is ranking the same list. HN's candidates have declared intent publicly, they carry independent proof of their work, and HA2 can't index them. Run both: HA2 for volume, the HN thread for senior candidates who sit outside that ranking.

FAQ

Is Hacker News "Who wants to be hired" worth sourcing from in 2026?

Yes, if you're hiring senior engineers and can offer remote work or relocation. The partial October 2026 count had 566 candidates against 221 hiring companies. 57 to 63% of candidates describe themselves as senior, and every one has opted in publicly with an email address. It's a weak channel for junior roles, which make up only 2 to 5% of postings.

Can recruiters email candidates from the HN thread?

Yes, as long as you're writing about a real work opportunity. The thread's rules say agencies and recruiters are off topic as posters, and ask readers to "only email these addresses to discuss work opportunities." Personal, role-specific emails are fine. Bulk sequences, network-building pitches and resume harvesting are not.

Where else can I find software engineers open to work?

LinkedIn's headline signal is thin. In Refolk's index, only 198 of 531,827 US software engineers write "open to work" in their headline, and only 11 do in Germany. Most intent on LinkedIn is hidden in private settings. Public opt-in channels like HN's monthly thread, combined with cross-referencing GitHub and personal sites, surface candidates the headline search misses.