The team is developing a digital assistants platform and an ecosystem of AI agents for bank employees. The solutions help automate processes, optimize costs, and implement generative neural networks into the business practices of the large and medium-sized segments.** The project is a new GenAI platform for building and industrial use of digital employees based on open-source LLM/VLM.

Company: Selecty (Recruitment Agency) 💸Income Level: 250-340K RUB net Format: Employment contract according to the Labor Code of the Russian Federation

⚙️ Tech Stack

Core: Python, FastAPI LangChain / LlamaIndex / Haystack PyTorch

Infrastructure / Ops: Kubernetes, Docker CI/CD (Jenkins, DevOps Kit) PostgreSQL, Redis, Tarantool S3

LLM / Inference: vLLM, Triton, SGLang RAG pipelines Vector and Graph databases

📊 Responsibilities

Development and enhancement of CI/CD pipelines for building and deploying AI agents Deployment and maintenance of LLM/VLM in production Optimization of inference and resource management in Kubernetes Design of backend services for the GenOps/self-service platform (REST, gRPC) Ensuring scalability and fault tolerance of LLM services Configuration of comprehensive monitoring (infrastructure + generation quality) Implementation and support of RAG scenarios Troubleshooting (3rd line support - a plus)

⚡️ Mandatory Requirements

Commercial experience in MLOps / DevOps for 3+ years Proficient work with Kubernetes (deployment, scaling, resource configuration) Experience in building CI/CD (Jenkins or similar tools) Deep understanding of Docker and containerization Experience with PostgreSQL, Redis Practical experience implementing LLM-based services Understanding of the ML stack and model principles (classical ML + LLM) Experience with PyTorch Skills in working with the LLM ecosystem (LangChain, etc.) Understanding of RAG approaches

➕ Will be a plus

Experience optimizing inference for large language models Work with vLLM / Triton / SGLang Experience with vector and graph databases Understanding of ML System Design Experience supporting highload systems Banking experience (not required)

💎 Conditions

Employment contract according to the Labor Code of the Russian Federation Voluntary Medical Insurance (including dentistry) Fitness compensation Discounts on English language learning (Skyeng) Salary indexation Modern equipment Company IT accreditation