ai-engineering AI Engineering Path
Start from LLM applications, Agent/MCP, RAG, real-time voice, model serving, and observability.
For backend-capable builders who want to turn model capabilities into runnable systems.
wzk.icu · Gleam Lab
AI Agent EngineeringMultimodal InteractionAI Application InfraBackend Platform
Zikang Wu | AI Agent × Multimodal Interaction × Backend Platform
He specializes in integrating LLMs, voice pipelines, business APIs, realtime communication, model services, and edge devices into deployable, measurable, and maintainable AI application systems.
Zikang Wu works across Java backend platforms, microservices, data links, high-concurrency systems, performance optimization, and engineering delivery, and applies that foundation to AI Agent systems, model services, realtime voice interaction, business intelligence, edge-cloud collaboration, and embodied-AI application systems.
655 Technical Notes Bilingual engineering archive
7 Project Stories AI, indie tools, automation
4 Focus Tracks Backend, AI, tools, writing
SSG Search Ready Sitemap, RSS, llms.txt and schema
Focus Areas
Backend Platform & Data Links
Java / Spring Cloud, microservice governance, Kafka / Redis / ES, Kubernetes / DevOps, reliability, and high-throughput systems
AI Agent & Model Services
LLM / ReAct / Function Call, MCP, vLLM, local model services, tool-call evaluation, and business API orchestration
Multimodal & Edge-Cloud
Wake word, VAD, ASR, LLM, TTS, WebSocket / gRPC, robots, patrol vehicles, and edge terminal integration
Writing & Knowledge Base
Long-term public writing on Java, big data, DevOps, AI Agent systems, model services, and engineering practice
Now Building
Currently Building
A compact view of the active workbench: public tools, AI engineering, knowledge systems, writing rules, and the site itself.
Experimental Project AscendLab
Organize 147 free browser tools and keep AI workflows clearly marked as preview, early access, or planned.
Public Tool Site In Development AI Voice Module
Continue Trace, latency, Apply / Reload, tool-list consistency, and end-to-end stability work.
Project Archive In Progress AI Infra Handbook
Build the map for model serving, RAG, Agent/MCP, GPU, Kubernetes, observability, and cost governance.
Learning Handbook Long-term Maintenance stop-slop-zh
Expand rule categories, prompt templates, review workflows, and possible Skill packaging.
Open-source Project Long-term Maintenance wzk.icu
Keep improving content governance, bilingual paths, SEO/GEO, and AI-search readability.
Public SiteProject Map
View All →wzk.icu - Personal Technical Portal
Long-term MaintenanceSolves: Turns a personal homepage into a maintainable, searchable, and AI-readable technical brand portal.
A personal technical portal rebuilt with Astro 6, combining blog publishing, project pages, personal branding, static generation, Markdown content, SEO, and AI-search-friendly structure.
AstroTypeScriptMarkdown
AscendLab - Indie Tools and AI Workflow Lab
Experimental ProjectSolves: Turns scattered tool ideas into accessible, maintainable, and reviewable indie product experiments.
AscendLab is an experimental site for browser-side tools, AI workflows, and indie building, exploring how small useful tools and automation flows can become maintainable public works.
Next.jsReactTypeScript
AI Voice Module - Edge-Cloud AI Interaction System
In DevelopmentSolves: Moves voice interaction from a model demo toward a configurable, observable, and extensible engineering system.
An edge-cloud AI voice interaction system experiment for multiple device carriers, exploring how STT, LLM, TTS, agent tools, runtime configuration, and observability form a low-latency maintainable voice pipeline.
PythonRustWebSocket
AI Infra Handbook - Learning and Practice Notes
In ProgressSolves: Turns scattered AI infrastructure knowledge into a durable, reviewable, and reusable engineering learning system.
AI Infra Handbook is a long-term learning and practice project around AI infrastructure, model serving, vector retrieval, GPU resources, Kubernetes deployment, observability, and cost-aware engineering.
AI InfraLLMGPU
stop-slop-zh - Chinese AI Writing Anti-Slop Rules
Long-term MaintenanceSolves: Turns fuzzy Chinese AI writing quality issues into a checkable, reusable, and collaboratively maintainable rule list.
stop-slop-zh is a rule project for Chinese AI writing quality, collecting common AI-flavored phrases, redundant sentence patterns, vague slogans, and mechanical transitions.
MarkdownChinese WritingAI Writing
Code Generator - Engineering Productivity for Business Development
Internal PracticeSolves: Turns repetitive business development from copy-paste work into a configurable, reviewable, and continuously improvable productivity workflow.
A code generator practice project for business development workflows, exploring how templates, metadata configuration, and AI assistance can generate common business code with less repetition and more consistency.
PythonFlaskJava
TeslaMate Automation Dashboard
Digital Life ExperimentSolves: Turns vehicle telemetry from passive records into an observable, alertable, and reviewable personal data system.
A Tesla vehicle data dashboard built on TeslaMate, Grafana, InfluxDB, and PostgreSQL, with trip records, charging analysis, energy metrics, alerts, and weekly reports.
DockerTeslaMateInfluxDB
Start Here
Reading Paths
Pick a track first, then follow the most relevant articles and projects without getting lost in the full archive.
ai-infra AI Infra Path
Understand the infrastructure behind AI apps through model serving, RAG, vector databases, GPU, Kubernetes, observability, and cost governance.
For backend, platform, DevOps, and AI app engineers who need an AI infrastructure map.
backend-engineering Backend & Platform Path
Build the engineering foundation around Java, Spring Cloud, microservices, Kubernetes, DevOps, messaging, and data pipelines.
For backend developers, platform engineers, and builders moving from business features to system design.
indie-building Indie Building Path
Understand how long-term public work grows through a technical portal, AscendLab, tool sites, SEO/GEO, AI workflows, and content systems.
For people building public work, personal brand sites, small tools, or indie product experiments.
ai-writing-quality Chinese AI Writing Quality Path
Use stop-slop-zh, anti-slop rules, technical-blog quality checks, review workflows, and prompt engineering to govern Chinese AI writing.
For writers, reviewers, prompt builders, and anyone reducing the templated feel of AI-generated Chinese content.
digital-life Long-term Records Path
Follow annual reviews, project write-ups, training notes, digital-life experiments, and personal reflections as long-term practice.
For readers interested in long-term growth, public notes, review habits, and practice beyond pure technical topics.
Topics
Topic Matrix
Six long-term topic assets connect Start Here, Series, Projects, Workflow, and the content roadmap.
backend-engineering Backend Engineering & Microservices
Turning business capabilities into stable systems with Java, microservices, databases, messaging, Kubernetes, and observability.
ai-engineering AI EngineeringTurning LLMs, Agents, MCP, RAG, real-time voice, and model APIs into runnable, observable, and iterative application systems.
ai-infra AI InfraBuilding the AI engineering foundation around model serving, RAG, vector databases, GPU, Kubernetes, evaluation, observability, and cost governance.
indie-building Indie Building & Tool SitesTurning a personal technical portal, tool site, content system, SEO/GEO, AI workflows, and public projects into long-term assets.
ai-writing-quality Chinese AI Writing Quality GovernanceReducing templated Chinese AI writing through stop-slop-zh, rule categories, common slop patterns, human review, and possible Skill packaging.
digital-life Long-term Records & Digital LifeTurning project reviews, personal data, digital life, annual records, and long-term growth into a public archive that can be revisited.
Featured Writing
Curated entry points into the main tracks, rather than a simple list of latest posts.
AI Engineering LLM, Agent, RAG, MCP, and model-serving practice.
LLM Application Engineering: Key Practices from Demo to Production
Core experience moving LLM applications from prototype to production: context management, error handling, cost control, observability. No basics, just real pitfalls.
Real-Time Voice & Robotics ASR/TTS, VAD, voice pipelines, and device interaction.
Real-time Voice Interaction Pipeline Latency Optimization
When building voice interaction systems, latency is the core experience metric.
Java / Microservices Java backend, messaging, service governance, and platform engineering.
Hive Slowly Changing Dimension Type 2: Order History State Management
Offline data warehouse needs to save order history state at low cost while supporting daily rollback and change analysis.
Data Engineering Hadoop, Hive, Kafka, Flink, and warehouse practice.
MQ Application: Cache Warm-up, Rate Limiting, Redis Lua and Traffic Peak Shaving
E-commerce seckill/ticket-grabbing scenarios with instantaneous traffic peaks, high read/write concurrency. Use pre-static + rate limiting queuing
Who This Helps
This site is built for practical readers
It is not a generic resume page. It is a growing archive of systems, tools, experiments, and writing paths.
Backend engineers moving into AI Engineering
You can follow how service design, observability, deployment, and tool calling connect with LLM applications.
Builders of indie tools and public products
You can inspect how small tools, project pages, SEO/GEO, and long-term writing form a public work system.
Readers who want practical technical archives
You can use the series, tags, project map, RSS, and llms.txt as structured entry points rather than a linear blog feed.
About
AI Agent, multimodal interaction, and backend platform engineer focused on complex systems integration and engineering delivery.
I integrate LLMs, realtime voice pipelines, business APIs, model services, and edge devices into deployable, measurable, and maintainable AI application systems.
Learn More →Public profiles
Personal Navigation Hub
Use wzk.icu as the fastest route to my code, writing, community profiles, and career page.