
We're looking for a proactive Middle AI / Agent Engineer to take full ownership of AI initiatives and build and implement advanced solutions that support real business workflows. Someone who treats LLMs as engineering components, not a cha…
richbrains · На сервисе с: 21.09.26 10:59
We're looking for a proactive Middle AI / Agent Engineer to take full ownership of AI initiatives and build and implement advanced solutions that support real business workflows. Someone who treats LLMs as engineering components, not a chat interface. You'll take an idea and turn it into something that runs in production: agents, RAG pipelines, prompt systems, integrations that solve real business problems.
No model training. Just solid, reproducible AI solutions built by a practitioner who thinks systematically and knows how to get consistent output from LLMs.
Responsibilities:
Write, test, validate, and iteratively improve prompts and system instructions until results are stable and reproducible;
Build and configure AI agents - tool-use, memory, multi-step workflows;
Implement RAG pipelines and connect LLMs to external data sources and APIs;
Write integration and automation code (Python, Node.js, or any scripting language);
Evaluate AI output quality in a structured way - accuracy, reliability, security and cost;
Compare models (OpenAI, Anthropic, Gemini etc) and recommend the right one for the task;
Monitor solutions in production, catch failures, and improve performance;
Stay current with LLM ecosystem - new models, tools, and frameworks - and apply relevant findings to your work.
Requirements:
Hands-on experience with LLMs and prompt engineering –systematic work, not occasional ChatGPT usage;
Experience building and configuring AI agents (planning, tool-use, memory);
Working knowledge of LLM APIs - OpenAI, Anthropic, Google, or similar;
Ability to write code in at least one scripting language for automation and integration tasks;
Understanding basic AI evaluation techniques;
Strong analytical and problem-solving mindset with attention to detail – able to explain why a solution was chosen;
English B2+ (written and spoken).
Nice to have:
Experience with RAG, vector databases and other retrieval/ranking techniques;
Multi-agent orchestration - LangChain, LangGraph or similar;
No-code / low-code automation experience;
Experience with STT / TTS integrations;
Understanding of MCP and agent protocols;
Cloud AI deployments - Amazon Bedrock or Azure OpenAI;
Familiarity with monitoring and observability tools for AI systems.
Working with us means:
Collaborating with a highly motivated and professional team, which values your ideas and expertise;
Regular performance reviews;
Comfortable office in Minsk, 5 minutes from the subway;
Working hours: 11:00–20:00 GMT+3;
Corporate discount programs;
Opportunities for training and certifications;
Team events and get-togethers to foster team spirit.
Join us!
*Данная вакансия является планируемой к созданию (перспективной).
HireSeeker собирает вакансии со всех площадок и присылает только релевантные. Бесплатно.