Hire AI Developer Intern: skills, stipend and screening

The three that matter most are Defining what "working" means before building; Reading an existing codebase before changing it; Ethics, bias and data provenance. Beyond those, look for working familiarity with Python, OpenAI or Anthropic APIs, scikit-learn or PyTorch. Everything else on the list above is teachable inside a term — treating it as an entry requirement shrinks your pool without improving the hire.

Ship one prototype with a measured success rate rather than a demo. That is sized for eight to twelve weeks of supervised work by someone with the fundamentals and no production experience. A second, smaller piece — ship one reviewed, merged change that goes to production during the term — usually fits alongside it. Anything requiring independent production judgement should stay with the reviewer.

Start from ₹19,000–₹47,500 a month, then adjust for city and duration: metros at the top, tier-2 typically 25–40% lower, and six-month commitments above six-week ones. Publish the number in the listing — "as per industry standards" is read as low or undecided, and it costs you applications from exactly the candidates who had another option.

Ask "How would you know this AI feature is working in production?" — you are listening for evaluation thinking — the difference between an engineer and a demo builder. Then follow the example they give rather than moving on to your next question. Score every candidate on the same set so the shortlist stays comparable.

Realistically two to four hours a week of a competent person: a longer session early on, then short daily availability and a weekly review. Below that, the intern stalls and produces nothing you can use. Above it, you are doing the work yourself. That time is the true cost of the hire, and it is what the stipend line in your budget does not show.

The "AI" qualifier needs a measurable success definition before any build. Demos are easy in this space and production is not, so agree what "working" means in week one. In screening terms, that means adding one specific check: defining what "working" means before building.

Yes, and the difference is what you must provide rather than what they must have. A developer brief promises build work, so it needs repository access, a real ticket and a named reviewer from day one. If those are not ready, hire against the general brief instead — a developer intern with nothing to build leaves inside a month, and usually tells their campus why.

Specificity does most of the work. A listing that names the project, the tools and the deliverable filters itself, because candidates can tell whether they fit. Adding one screening question to the application — from the set above — removes most of the rest without adding a review round.

Always. Students plan around semester dates, and a listing without a start date and duration is filtered out by exactly the organised candidates you want. For Artificial Intelligence roles, stating "three months, starting June" typically produces more applications than an open-ended listing at a higher stipend.