From Translator to Prompt Engineer

This article was written by Uliana Andriiashchuk, CEO of KLS Translation Agency, following her participation in Meet Central Europe 2025 — a key event for language industry professionals across Central Europe.

It’s no secret anymore that the translation industry is changing at an incredible pace — almost at the speed of light. And it’s important to recognize this transformation.
Not long ago, we were talking about NMT (Neural Machine Translation), and now AI translation has become the new standard. The discussion is shifting toward evolving roles within the translation process. Will we soon forget what “classic translation” means and move entirely toward post-editing and adapting AI output? The next 5–10 years will show.

What drives this change?
Primarily, the rise of large language models (LLMs) — systems that can generate, translate, edit, and transform text into various formats. Translation, in this sense, becomes only one part of a much broader workflow.
Equally important is the growing ecosystem of tools and software that translators can integrate into their daily routine to accelerate productivity. These tools are rapidly improving and competing for efficiency.

That’s why for small and mid-sized translation companies, it’s now crucial to have people on the team who are willing to experiment with AI-based tools, test them, and implement them in everyday operations. Many professionals note that these tasks should be handled by dedicated specialists, rather than combining them with a translator or editor’s workload — a highly effective approach.

As a result, new hybrid roles are emerging — ones that would have sounded unusual just a few years ago.
One example is the prompt engineer — a new, highly specific role that has appeared with the development of LLMs and AI platforms.

What’s their mission? To ensure that AI performs tasks as accurately and effectively as possible.
A prompt engineer crafts and refines prompts — carefully structured inputs that help the model produce relevant, precise, and useful results. They fine-tune wording, structure, context, and constraints so that AI “understands” the request correctly.
Not quite a programmer, and not quite a translator.

Organizations are still experimenting with titles for this position: some call it an AI specialist, others — an AI wizard.
But the title itself matters less than defining the core value this role brings to your company.