Java Developer (AI / GenAI)

Middle Java Developer for SberSpasibo, Moscow office. Develop microservices on Java 17+, build RAG pipelines and multi-agent systems using LangChain/LangGraph.

What you'll do: - Design and develop microservice architecture on Java 17+ using Spring Boot (MVC, Data JPA, Security) - Build Retrieval-Augmented Generation (RAG) pipelines: from parsing raw data to embedding indexing - Develop and support Qdrant vector databases: design sharding schemas, configure hybrid search (BM25 + Dense Vectors), optimize vector queries - Create multi-agent systems based on LangChain/LangGraph to automate business processes (classification, entity extraction, decision making) - Deep prompt engineering: write prompts from scratch for different use cases, implement Few-shot, Chain-of-Thought (CoT), structured output (JSON mode) - Debug model semantics: reduce answer entropy, manage variability via inference parameters (temperature, top_p, penalties), implement self-healing mechanisms for degradation of external LLM providers - Optimize PostgreSQL performance: tune complex SQL queries, work with indexes (B-tree, GIN/GIST, trigrams), design transaction isolation - Move computations into container environment: Docker, Kubernetes (Helm), build fault-tolerant CI/CD pipelines in GitLab CI/Jenkins

What we expect: - Strong Java 17+, understanding of multithreading (CompletableFuture, virtual threads Project Loom is a plus) - Expert level in Spring stack: deep understanding of bean lifecycle, AOP, fine-tuning of DB connection pools and security parameters - Experience in designing relational models in PostgreSQL, ability to read EXPLAIN ANALYZE, deal with locks, write efficient migrations (Liquibase/Flyway) - Practical experience with vector DBMS (Qdrant) or Weaviate/Pinecone/Milvus. Understanding of distance metrics (Cosine Similarity, Dot Product, Euclidean Distance) - Experience with LangChain and/or LangGraph libraries. Ability to build state graphs and integrate external tools (tool calling) - Prompt Engineering skills and managing LLM behavior without changing model weights - Experience with Docker and Kubernetes (manifests, Helm charts). Confident Linux shell user - CI/CD processes (GitLab CI or Jenkins): build with Maven/Gradle, test coverage (JUnit 5, Testcontainers), static code analysis

Full-time employment, office format in Moscow.