AI Prompt Engineer

SoftSvit

  • Киев
  • Постоянная работа
  • Полная занятость
  • 23 д. назад
We are expanding our team and looking for a Prompt Engineer to join our RnD department. You will be responsible for designing, testing, and optimizing prompts for existing Large Language Models (LLMs) that process large volumes of text and audio data. These models are used for summarization, prioritization, tone analysis, and extracting key information.The role involves analyzing dataset outputs, identifying failure cases, refining prompts, and quickly iterating based on internal feedback and testing. Optionally, you may work with Python or PHP scripts to support dataset exploration and evaluation.Who We AreOur RnD department is a key driver of innovation in the company, consisting of ten teams that develop CRM, marketing tools, CMS, payment solutions, and internal systems. With a strong focus on automation, performance, and scalability, we create solutions that enhance efficiency across multiple business units.Why Join Us?
  • You'll work directly with LLM-powered systems that automate summarization, classification, and prioritization tasks across high-volume content flows.
  • You'll be part of a process where prompt design directly impacts product output and user experience.
  • You'll collaborate with a highly skilled RnD team - including developers, analysts, QA, DevOps, and product managers.
  • We promote experimentation, knowledge sharing, and give you the freedom to test new ideas and approaches.
  • If you enjoy working with AI tools and want to make them smarter and more reliable-this role is for you!
RequirementsMust-Have:
  • 1+ year of hands-on experience with LLMs like ChatGPT, Claude, or similar frameworks.
  • 2+ years of experience in a technical or analytical role (e.g. technical writer, analyst, QA, developer).
  • Understanding of how GPT-style models behave and fail in real-world usage.
  • Experience with SQL/PostgreSQL and data querying.
  • Ability to read, build, and debug scripts on Python or PHP.
  • Strong logical thinking and excellent written communication skills.
Nice-to-Have:
  • Completed a course or built a practical project in prompt design or LLM application.
  • Familiarity with prompt-based agents or RAG (retrieval-augmented generation) architectures.
Responsibilities
  • Analyze business tasks and develop high-quality prompts for LLMs that handle summarization, prioritization, data extraction, and tone adjustment.
  • Work with text and audio datasets, chunk datasets, test prompt variations and iterate based on results and team feedback.
  • Perform data analysis at scale, optionally using Python scripts for support.
  • Collaborate with developers, analysts, and product managers to deploy and evaluate LLM-based tools.
  • Research and document LLM behavior, limitations, and failure patterns to improve prompt strategies.
  • Ensure prompt outputs are accurate, consistent, and aligned with project goals.
Benefits
  • 18 business days of paid vacation + public holidays compensation.
  • Insurance Fund of the company.
  • Partial coverage of all professional studies.
  • Coverage of sick leaves, sports activities, and English language courses.

Rabota