Описание:
The company builds AI solutions for enterprise clients. Its systems run on self-hosted infrastructure, including open-source LLMs, on-premises servers, and private cloud.
Задачи:
- Design and build production RAG systems and AI agents
- Run open-source LLMs on self-hosted infrastructure, from model selection to serving and optimization
- Build retrieval pipelines for document parsing, chunking, embeddings, hybrid search, and reranking
- Improve system reliability through evaluation, guardrails, monitoring, and hallucination control
- Deploy systems on Kubernetes with DevOps engineers
- Work directly with client teams from requirements gathering to production rollout
Требования:
- 4+ Years of professional Python development
- At least one AI/LLM system in production
- Hands-on experience with LangChain and/or LangGraph
- Experience with vector databases such as pgvector or Qdrant
- Experience with self-hosted LLMs such as vLLM, Ollama, or TGI
- Docker and basic Kubernetes skills
- English at B2 level or higher
- Working hours overlapping with UTC+0 to UTC+4
- Registered as self-employed or with your own company and able to issue invoices
- Будет плюсом: RAG evaluation (RAGAS, Langfuse, or similar), TypeScript / Node.js / React or Java, on-premises or private cloud deployments
Условия:
- Freelance, project-based work
- Long-term collaboration opportunities
- Challenging AI projects that support professional expertise and technical skills development
- Opportunities to work with modern AI technologies and solve real-world engineering problems