Your students can code. That is no longer enough to get hired.
Employers now want proof a graduate can work on a team, and fewer junior jobs reliably provide it. Blacksmith adds that experience, without changing anything in your programme.
The rung your students used to step onto is thinning out.
Placement is the promise every programme makes: to students, to parents, to funders. It was never really the curriculum that kept it. It was the job that came next.
You teach them to build.
Your graduates leave able to write software. That part of the programme works, and it is not what is failing them.
The junior job was the bridge.
It turned people who could code into engineers, by assigning them work, reviewing it, and vouching for them afterwards.
That bridge is getting narrower.
AI handles more of the simple work juniors used to learn from, while leaner teams ask graduates to arrive with proof of judgement already attached.
Early-career workers aged 22–25 in the most AI-exposed occupations saw a 16% relative employment decline after generative AI adoption.
Stanford Digital Economy Lab, 2025 ↗The New York Fed major table reported 6.1% unemployment for recent computer science graduates and 7.5% for computer engineering graduates.
Federal Reserve Bank of New York, 2025 ↗New-grad hiring at the 15 largest tech firms is down more than half since 2019. New grads are now about 7% of their hires.
SignalFire State of Tech Talent, 2025 ↗They join a real engineering team.
Eleven AI specialists (a manager, a tech lead, reviewers, a tutor for whatever they got stuck on) running real work cycles on code the student owns.
Eleven AI specialists (a manager, a tech lead, reviewers) running real work cycles on code the student owns.
Assigned tasks, pull requests, code review and kickoff calls. The actual rhythm of a job, not a simulation of one.
Whoever flags a gap in a student’s work teaches it, by message or by voice, using their own code as the example, and only marks it done once the student can explain it back.
A verified Level 1 to 5, tied to the specific work that earned it, on a GitHub history any employer can open.
The difference between a graduate who has read about code review and one who has survived forty rounds of it.
You change nothing in your programme.
No module to rewrite, no timetable to move, no staff to retrain, and less on your instructors than before, because the repetitive half of teaching moves to the team. Give every student a senior engineer, and give every instructor their evenings back.
It runs alongside your course.
Or after it, as the step that used to come from a first job. No module to rewrite, no timetable to move, no staff to retrain.
Students join by one link.
Enrol a class, a term or a whole programme at once. Onboarding is a link and a first conversation, not a setup project.
Start with one cohort.
Run a pilot on a single intake, read the levels and the placement signal at the end of it, and decide from evidence.
Your instructors stop being the bottleneck
Every student has a teammate who will explain the same concept at 1am for the fortieth time without tiring. Your staff spend their hours on curriculum, on judgement, and on the students who need a person.
Cohort seats
Enrol students in groups, by class, term or programme. Onboarding is one link, not a setup project.
An institution dashboard
Every student’s level, progress and the evidence behind it, so you can see who is moving and who is stuck.
Co-branded credentials
Each graduate gets a shareable profile employers can verify, carrying your name alongside ours.
Outcomes data you can use
Placement signals and competency levels you can put in front of prospective students and funders.
Run it with one cohort first.
We bring the engineering team; you bring the students. Tell us the size of your next intake and when it starts.