Projects | 1byteone - Java Backend / AI Application Developer

LingXi AI Writing Platform

Mon, 24 Aug 2026 00:00:00 +0000

AI-powered intelligent article creation platform built on Spring Boot 3 + Spring AI Alibaba, independently designed and developed full-stack.

Project Overview

An AI-powered intelligent article creation platform built on Spring Boot 3 + Spring AI Alibaba. As an independent full-stack project, it explores the practical application of LLMs in long-form content generation.

The system innovatively introduces a StateGraph workflow engine to orchestrate multi-agent collaboration, combined with SSE streaming output, solving the problems of long wait times and loose content structure in traditional AI generation. The system implements a complete automated loop from inspiration to image generation.

Core Features

Feature	Description
🧠 Multi-Agent Orchestration	StateGraph workflow engine for flexible orchestration of title, outline, and body generation agents
⚡ Streaming Output	SSE technology for millisecond first-token display, eliminating wait anxiety
🔄 Async Task System	ThreadPoolTaskExecutor + @Async decouples long-running operations
🖼️ Multi-Source Images	Strategy pattern for dynamic Pexels/SVG image service integration with auto-fallback
📐 Structured Output	Prompt template engine + StructuredOutputConverter for unified LLM output format
📊 Task Tracking	Progress management and result callback mechanisms

Technical Highlights

1. Multi-Agent Orchestration Architecture

StateGraph-based workflow engine replaces traditional linear calls
Title, outline, and body generation agents in serial/parallel combinations
New generation modes require only new node configuration

2. High-Concurrency Streaming Response

SSE (Server-Sent Events) integration with frontend
ThreadPoolTaskExecutor async task system
Millisecond first-token display, 50% throughput improvement

3. Fault-Tolerant Generation Pipeline

Prompt template engine + StructuredOutputConverter for unified output
Strategy pattern for multi-source image services
Auto-fallback on single source failure, 99%+ success rate

4. Async Task & Status Tracking

@Async decouples article/image generation as background tasks
Task status tracking with progress management
Significantly improved system throughput

Project Results

Architecture: StateGraph workflow engine + multi-agent orchestration
Performance: Millisecond first-token display, 50% throughput improvement
Reliability: Multi-source image auto-fallback, 99%+ success rate
Extensibility: New generation modes require only new node configuration

Project Status: ✅ Completed
GitHub:

ZhiYi Elderly Care Nursing System

Mon, 24 Aug 2026 00:00:00 +0000

Comprehensive management platform for medium-sized nursing homes (B-end ops dashboard + C-end family mini-program), covering health assessment, IoT device monitoring, and care work orders.

🔒 Repository is not yet public. Project details below.

Project Overview

Comprehensive management platform for medium-sized nursing homes with both B-end operations dashboard and C-end family mini-program. Covers health assessment, IoT device monitoring, and nursing work orders.

Team of 6 developers; responsible for the health assessment engine, IoT sync service, and unified authentication modules, serving 500+ devices and 2000+ daily family visits.

Core Responsibilities

1. Health Assessment Module Development & Performance Optimization

Integrated Baidu Qianfan AI model for semantic analysis of health reports
Designed Redis multi-level caching for frequently accessed health advice
Reduced single analysis response from 8s to under 1s
Improved nurse review efficiency by ~30%
Effectively supported high-concurrency queries during peak morning hours

2. IoT Device Management Module Design

Developed Huawei Cloud IoT platform integration service
Incremental sync of product lists via async thread pool
“Device-Location” binding logic for data consistency
Sync time reduced from 3 minutes to 10 seconds
Real-time monitoring of 500+ devices

3. Security Authentication System

Interceptor + ThreadLocal based seamless login for WeChat mini-program users
Multi-thread data isolation and security
Zero memory leak stable operation
Supporting 2000+ daily active visits

Project Results

Performance: AI analysis response reduced from 8s to under 1s
Sync: IoT device list sync from 3min to 10s
Scale: 500+ devices real-time monitoring, 2000+ daily visits
Stability: Zero memory leak, stable production operation

Project Status: ✅ Completed
Repository: 🔒 Not yet public — coming soon

Distributed Micro Cloud Mall

Fri, 15 May 2026 00:00:00 +0000

Microservice E-commerce Platform - SpringCloud Alibaba Stack (Nacos/Gateway/Sentinel) + MyBatis-Plus + Feign Remote Calls.

Project Overview

Complete practice from monolithic to microservice architecture, covering enterprise middleware integration and distributed solutions. The project benchmarks against mid-senior Java backend engineer core competency models.

Professional GOAL (Position Capability Benchmarking)

Level	Capability Domain	Coverage Modules	Weight
🥇 Distributed Architecture	Microservice splitting, service governance, distributed transactions, message queues	mall-order-service, mall-seckill-service, Middleware	35%
🥈 Middleware Depth	Redis cache, ES search, RocketMQ messaging, Seata transactions, Sentinel rate limiting	docs/Middleware/ (11 middlewares)	30%
🥉 Business Implementation	Product/User/Cart/Order CRUD, JWT auth, AOP logging	mall-product-service, mall-user-service, mall-cart-service	25%
🏅 Engineering	Docker containerization, Gateway, Nacos config center, CI deployment	mall-gateway, docs/docker/	10%

Technical Architecture

┌──────────┐ ┌──────────┐ ┌──────────┐
│ Client │ │ Client │ │ Client │
└────┬─────┘ └────┬─────┘ └────┬─────┘
 └──────────────┴──────────────┘
 │
 ┌──────┴──────┐
 │ Spring Cloud │
 │ Gateway │ ← Unified Auth + Routing + Rate Limiting
 └──────┬──────┘
 │
 ┌──────────┼──────────┐
 │ │ │
┌────┴────┐ ┌──┴───┐ ┌───┴────┐
│ Services │ │Middle│ │Infra │
│ User │ │Redis │ │Nacos │
│ Product │ │MySQL │ │Sentinel│
│ Order │ │Mongo │ │Seata │
│ Cart │ │ES │ │XXL-Job │
│ Pay │ │MQ │ │ │
│ Seckill │ │Canal │ │ │
└─────────┘ └──────┘ └─────────┘

Project Structure

mall-micro-cloud/
├── mall-api/ # Feign Interface Layer
├── mall-common/ # Common Utilities
├── mall-gateway/ # Spring Cloud Gateway
├── mall-services/
│ ├── mall-user-service/ # User Service (JWT + BCrypt)
│ ├── mall-product-service/ # Product Service (MyBatis-Plus)
│ ├── mall-cart-service/ # Cart Service (Redis Cache)
│ ├── mall-order-service/ # Order Service (Seata Distributed Tx)
│ ├── mall-pay-service/ # Payment Service (RocketMQ Tx Messages)
│ ├── mall-seckill-service/ # Seckill Service (High Concurrency)
│ ├── mall-es-service/ # Search Service (Elasticsearch)
│ ├── mall-oss-service/ # Object Storage Service
│ ├── mall-scheduler-service/ # Scheduler Service (XXL-Job)
│ ├── mall-consumer-service/ # Message Consumer Service
│ └── mall-test-service/ # Test Service
├── docs/
│ ├── Middleware/ # 11 Middleware Tutorials
│ ├── interview/ # Interview Questions (9 chapters)
│ ├── db-sql/ # Business SQL
│ └── docker/ # Docker Deployment Guide
└── pom.xml # Maven Parent POM

Technical Highlights

1. Distributed Architecture

Microservice Splitting: Independent services by business domain
Service Governance: Nacos discovery + Gateway routing
Distributed Transactions: Seata AT mode for cross-service consistency

2. Deep Middleware Integration

Redis: Cache + Distributed Lock + Session Management
Elasticsearch: Full-text Search + IK Analyzer
RocketMQ: Async Messaging + Transaction Messages
Sentinel: Rate Limiting + Circuit Breaking

3. High-Concurrency Seckill

Redis pre-deduction + Redisson reentrant lock
RocketMQ async persistence prevents overselling
Sentinel rate limiting protects system stability

4. Engineering Practices

Docker containerization
Gateway unified authentication
Nacos configuration center

Project Results

Architecture: 11 microservices designed and governed
Middleware: 11 enterprise middlewares integrated
Performance: 10k+ concurrent stress testing passed
Documentation: Complete middleware tutorials and interview questions

Project Status: ✅ Completed
GitHub:

Agri-QA-Assistant

Fri, 20 Mar 2026 00:00:00 +0000

Agricultural Intelligent Q&A Prototype System based on LangGraph Goal-Oriented Agent Architecture.

Project Overview

Agri-QA-Assistant is a production-grade prototype for agricultural knowledge retrieval and question answering. It combines Retrieval-Augmented Generation (RAG) with a goal-oriented agent architecture to provide accurate, context-aware answers about crop cultivation, pest management, fertilization, and agricultural policy.

Core Features

Feature	Description
🌱 Domain-Specific RAG	ChromaDB vector store with curated agricultural knowledge base covering crops, pests, fertilizers, soil, and machinery
🧠 Multi-turn Memory	SQLite-backed conversation history with context continuity across sessions
🎯 Intent-Aware Routing	LangGraph agent routes queries to RAG, general knowledge, or tool-augmented paths
📊 Evidence Grounding	Citation-backed responses with source attribution and faithfulness scoring
🔧 MCP Integration	Open MCP servers for web fetch, temporal queries, and extensible tool use
🎨 Apple Liquid Glass UI	Frosted glass effects, translucent layers, iOS-style animations with Tailwind CSS

Technical Architecture

┌─────────────────────────────────────────────────────────────────┐
│ Frontend: Next.js 14 + Radix UI │
│ Apple Liquid Glass Chat Interface │
│ ┌─────────────┐ ┌──────────────┐ ┌─────────────────────────┐ │
│ │ Chat Panel │ │ Knowledge │ │ Generative UI │ │
│ │ + Streaming │ │ Panel │ │ (Crop Diagnosis, etc.) │ │
│ └─────────────┘ └──────────────┘ └─────────────────────────┘ │
└────────────────────────────┬────────────────────────────────────┘
 │ HTTP / SSE
┌────────────────────────────▼────────────────────────────────────┐
│ Backend: FastAPI + LangGraph │
│ ┌──────────────┐ ┌──────────────┐ ┌────────────────────────┐ │
│ │ Intent Router │ │ RAG Pipeline │ │ Tool Executor │ │
│ │ (LangGraph) │ │ (ChromaDB) │ │ (MCP Servers) │ │
│ └──────────────┘ └──────────────┘ └────────────────────────┘ │
└─────────────────────────────────────────────────────────────────┘

Technical Highlights

1. Goal-Oriented Agent Architecture

Built on LangGraph’s agent framework, the system dynamically routes queries based on user intent:

RAG Path: Retrieves information from agricultural knowledge base
General Knowledge Path: Calls general LLM for answers
Tool-Augmented Path: Integrates external tools for real-time data

2. Evidence Grounding & Faithfulness Scoring

Each response includes source citations
Faithfulness scoring ensures information reliability
Users can verify response sources

3. Apple Liquid Glass UI

Modern frosted glass effect interface
iOS-style animations and interactions
Responsive design for multi-device support

4. MCP Server Integration

Open tool-calling architecture
Supports web fetch, temporal queries, and extensions
Pluggable tool ecosystem

Project Results

Accuracy: 95%+ agricultural knowledge Q&A accuracy
Usability: Multi-turn conversation context retention
Experience: Modern Apple Liquid Glass UI
Extensibility: MCP tool integration architecture

Project Status: ✅ Completed
GitHub: