Remote job
Middle AI va LLM muhandis
Job details
About this role
Role overview Build retrieval-augmented generation pipelines, AI assistants, and agents using open-source and cloud-based large language models. The work spans implementation and production delivery, including document handling and assessing the quality of generated answers.
Responsibilities - Develop RAG pipelines and AI assistants or agents with Python. - Work with commercial LLM APIs and open models. - Implement retrieval using embeddings and vector search, and design prompts. - Process source documents, including PDFs and OCR inputs. - Evaluate answer quality and take features from design through production.
Requirements - 2–4 years of commercial experience in a relevant field. - Python experience and familiarity with the LLM ecosystem, including APIs such as OpenAI or Anthropic and open models. - Understanding of embeddings, vector search, and prompt design for RAG systems. - Experience evaluating generated responses and working with documents such as PDFs and OCR output. - Ability to deliver features independently from design to production. - Strong written English; Russian or Uzbek is an advantage.
Benefits and work setup - Remote-first work, with the option to work from an office in Tashkent on a hybrid basis. - Projects include private LLM and RAG platforms, AI assistants, and enterprise systems for customers in Uzbekistan, South Korea, and the UAE. - Market-level compensation in USD, based on experience. - Work with a small senior team and report directly to the CEO.