How Gemini helps a non-programmer handle small-business IT tasks without freelancers

A manager of a small company, far from professional programming, described on Habr his experience using Gemini for everyday IT tasks. The conclusion is surprising: work that previously required payment to freelancers and integrators is now solved in a conversation with an AI — faster, cheaper, and without bureaucratic approval processes. Small businesses constantly face tasks that are too small for a dedicated IT department, but complex enough to be unsolvable without technical knowledge.

Parse data from a supplier's website, convert a file to the required format, fix button logic on a website — each of these takes a specialist 30–60 minutes but is often charged as two hours of mid-level work. Plus time spent finding a contractor, writing a technical specification, and reviewing the result. By his own account, the author is not a programmer, DevOps engineer, or 1C specialist.

He has basic familiarity with Python, experience working with WordPress and Joomla, and a general understanding of how parsing works. This proved sufficient to correctly formulate tasks for Gemini and verify its output. The key skill here is not knowledge of syntax, but understanding of business logic: what exactly needs to be done and why.

The specific tasks the author solved with AI, he intentionally does not call masterpieces. This is standard routine: data export automation, small scripts, website edits. By professional standards — tasks for a junior.

But the very fact that a person without relevant training can handle them independently changes the economics of small business: the contractor is no longer necessary where once unavoidable. The psychological aspect matters too. Previously, each such task meant a choice: pay and wait, or postpone.

Now there is a third option — do it yourself, asking AI to explain the steps and write the code. Gemini in this scheme acts not as a replacement for a programmer, but as a tutor who prompts and corrects while the non-programmer iterates toward the result. The author is honest in his assessments: he acknowledges that his story is not about perfect code.

It is a story about lowering the entry threshold. The AI makes small technical tasks accessible to people with business understanding but without programming background. It is to such an audience — managers, entrepreneurs, operational leaders of small companies — that such experience is most relevant.

The practical outcome is simple: if a company has someone with basic technical knowledge and willingness to experiment, Gemini and similar tools can already reduce spending on small contractors. Not because AI programs better than they do — but because the threshold beyond which a task can be done independently has become significantly lower.