Description
The team is responsible for maintaining the bank's automated systems in accordance with the service level agreement. The area of responsibility includes systems built on a wide variety of technology stacks: (Kafka/SDP Hadoop/Postgres/OpenShift (K8S)).
Responsibilities
- designing and implementing the logic for generative AI models, using frameworks (LangChain, LangGraph) to develop agent-based and multi-agent systems, performing predictive analytics and anomaly analysis based on system logs
- connecting to vector databases (PostgreSQL with the pgvector extension, redis, mongodb, etc.) for RAG architecture, developing APIs (REST) for agent integration
- writing pipelines and playbooks for Jenkins for various automation tasks, ranging from simple to quite complex comprehensive solutions. Technologies used: groovy, python, bash
- designing and building RAG pipelines (establishing the entire data lifecycle for RAG: from extraction via APIs (REST/GraphQL), web scraping (Scrapy), and parsing complex documents (PDF, DOCX) to fine-grained cleaning, segmentation into meaningful blocks, and enrichment with metadata)
- maintaining the designed and developed solutions
Requirements
- knowledge of high-level programming languages (Python, Java, etc.), understanding of OOP principles, algorithms, data structures, and software design.
- experience in developing and implementing solutions based on generative AI, including creating AI agents.
- skills in working with RAG methods (Retrieval-Augmented Generation), integrating LLMs, and optimizing generation accuracy
- experience in creating and optimizing vector stores, implementing intelligent search, working with embedding models; experience with various types of neural networks is a plus.
- basic experience working with Linux
- basic experience working with Kubernetes (K8S), Docker, Helm
- experience with DevOps tools, version control systems, and distribution repositories (Jenkins, BitBucket, GIT, Nexus, etc.). Experience obtaining data from various types of APIs (REST, GraphQL, SOAP)
Conditions
- comfortable modern office near Kutuzovskaya metro station (for the office at Kutuzovsky 32)
- office-based work format
- annual salary review and annual bonus
- corporate gym and relaxation areas
- flexible discount on mortgage loans, equal to 1/3 of the Central Bank's key rate
- more than 400 educational programs from SberUniversity for professional and career development
- onboarding program and supervisor assistance at the start
- extended voluntary health insurance (VHI), preferential insurance for family, and corporate pension program
- free SberPrime+ subscription, discounts on products from partner companies.
