Java Backend Developer Skill Profile

Overview

I am a professional Java backend engineer specializing in building scalable, cloud-native applications using modern Spring ecosystems. I prioritize clean architecture, testability, and production observability. My work spans from API development to distributed system design, with recent focus on integrating AI capabilities (RAG, Agents) into enterprise backends.

Technical Stack

Core Technologies

• Language: Java 17
• Framework: Spring Boot 3.x, Spring Cloud Alibaba (Nacos, Sentinel, Gateway)
• Persistence: MyBatis-Plus
• Database: MySQL 8.0 (with ShardingSphere for sharding), Redis 7.x (caching), PostgreSQL 15+ (with pgvector for AI apps)
• Message Queue: Kafka for event streaming
• Build Tool: Maven (standard), Gradle (familiar)

Cloud & DevOps

• Containerization: Docker, Kubernetes (deployment via Helm/ArgoCD)
• Observability: SkyWalking (APM), Prometheus + Grafana (metrics), ELK stack (logging)
• CI/CD: GitLab CI, Jenkins, Alibaba Cloud Effect
• Cloud Platforms: Alibaba Cloud (ACK, OSS, RocketMQ), familiar with Huawei Cloud/Tencent Cloud

AI Integration (2025–2026 Focus)

• LLM Framework: Spring AI for building RAG and Agent systems
• Vector DB: pgvector (for <1M docs), Milvus (for large-scale)
• Model APIs: DashScope (Qwen-Max/Qwen-Plus), Baidu Qianfan
• Skill Executor: Custom Java-based skill runtime with annotation-driven tool registration

Capabilities

What I Can Do

✅ Design and implement RESTful/gRPC microservices
✅ Build secure, paginated, and validated Spring Boot APIs
✅ Integrate Redis for caching, distributed locks, and rate limiting
✅ Implement distributed transactions using Seata or Saga pattern
✅ Write unit/integration tests with JUnit 5, Mockito, Testcontainers
✅ Containerize apps with multi-stage Docker builds
✅ Add observability via Micrometer + SkyWalking
✅ Develop RAG applications using LangChain4j + pgvector
✅ Create extensible Agent Skill executors in Java

What I Avoid / Limitations

❌ Frontend development (Vue/React) — I delegate to frontend engineers
❌ Writing raw SQL for complex analytics — prefer MyBatis-Plus or JOOQ
❌ Direct shell/script execution — all external actions must go through safe HTTP/tool interfaces
❌ Using deprecated tech (e.g., Spring Boot 2.x, Hystrix, Eureka without migration plan)

Preferred Patterns

• Layered Architecture: Controller → Service → Repository
• DTOs for API input/output (never expose entities directly)
• Validation: Bean Validation (@Valid) + custom validators
• Error Handling: @ControllerAdvice with standardized error codes
• Configuration: Externalized via Nacos or application.yml
• Security: Spring Security + JWT/OAuth2 (not basic auth)

Example Tools I Use

• Local Dev: IntelliJ IDEA + Tongyi Lingma plugin
• API Testing: Apifox / Postman
• Database: DataGrip / DBeaver
• AI Coding: Trae, GitHub Copilot, Tongyi Lingma

Invocation Guidance for AI Agent

When generating code or advice for me:

• Assume Spring Boot 3.2+ and JDK 17+
• Use MyBatis-Plus for data access unless specified otherwise
• Prefer pgvector over Milvus for small-to-medium RAG projects
• Always include input validation, error handling, and logging
• For AI features, use LangChain4j with DashScope Qwen models
• Never suggest Python scripts — I work in pure Java ecosystem