Gleam Lab · About

Personal Technical Profile

Hi, I'm Zikang Wu.

AI Agent × Multimodal Interaction × Backend Platform × Edge-Cloud Systems

I am a systems engineer grounded in Java backend platforms, microservice architecture, data links, and engineering delivery. I now extend that foundation into AI Agents, model services, realtime voice pipelines, multimodal interaction, edge-cloud collaboration, and embodied-AI application systems, turning complex business needs into deployable, measurable, and maintainable AI application systems.

About Zikang Wu page cover: backend engineering, AI Engineering, and indie building

Who I Am

I think of myself as a long-term builder: I write code, build systems, create tools, and keep records of the process. This website is not just a resume page. It is my public technical archive for articles, projects, experiments, and milestone reviews.

In one sentence, I am best understood as: An AI Agent, multimodal interaction, and edge-cloud systems engineer with around five years of backend platform and data-link experience. My core value is not single-point algorithm research, but integrating backend platforms, model services, business APIs, realtime communication, and edge devices into reliable business loops.

What I focus on is not "do I know this technology", but rather:

Public Capability Evidence

The following points are public-safe engineering evidence. Phone numbers, internal endpoints, customer data, production addresses, and unredacted logs are intentionally excluded.

Platform Scale 4,000+ communities / 20,000+ devices / 5B+ capture records

Helped evolve a smart-community data platform for capture ingestion, parsing, statistics, forwarding, display, and operations monitoring.

Data Throughput 8M+ daily data throughput

Supported high-frequency data links with Kafka async flow, compensation, cache governance, query optimization, and containerized delivery.

Performance Optimization Core APIs ~400ms -> 10ms, aggregation queries ~1000ms -> 20ms

Optimized query paths with Elasticsearch wide-table aggregation, Redis cache, data-structure tuning, and hot/cold data governance.

Realtime Voice Pipeline Wake word / VAD / ASR / LLM / tool use / TTS / playback loop

Integrated edge capture/playback, cloud model services, business API orchestration, and WebSocket/gRPC realtime communication.

Agent Evaluation Function Call test sets + EFK logs + human spot checks

Focuses on tool-call accuracy, parameter recognition, failure analysis, and business-action closure instead of demo-only effects.

Edge-Cloud Safety Boundary Environment recognition + LLM semantic judgment + local radar/control checks

Adds safety gates before robot and patrol-vehicle task dispatch, avoiding direct LLM-only control over embodied device actions.

What I Do

My core background is Java backend and platform R&D. I work extensively with Spring Boot / Spring Cloud Alibaba stack, with experience in:

Backend & Platform Engineering

  • Java / Spring Boot / Spring Cloud Alibaba
  • Microservice governance, gateway, auth, tracing
  • MySQL, Redis, Elasticsearch
  • Kafka, RabbitMQ, data pipelines
  • Docker, Kubernetes, Rancher, CI/CD
  • EFK, troubleshooting, performance optimization, delivery

AI Engineering

  • LLM application engineering and model service integration
  • Agent / MCP / Function Calling
  • ReAct, multi-turn parameter completion, business API orchestration
  • STT / LLM / TTS real-time pipelines
  • Trace / Observability / Runtime Config
  • Cost, latency, reliability, and evaluation governance

Indie Building

  • AscendLab and browser-side tools
  • AI Workflow previews and rule repositories
  • stop-slop-zh, code generators, automation scripts
  • Personal productivity systems and public work polish

Writing & Long-term Archive

  • Technical blog, project reviews, tutorial series
  • AI research notes, content governance, SEO / GEO
  • Annual reflections and digital-life experiments

In recent years, I've participated in building systems from scratch and done architecture refactoring, performance optimization, and platform evolution from 1 to N. Rather than just writing business APIs, I prefer tackling problems that are complex, messy, long-chain, and require real-world delivery: throughput, latency, reliability, tool use, permissions, logs, deployment, terminal integration, and user interaction loops.

Current Projects

Public Boundary

This page explains capability, direction, and engineering method without publishing a full resume or internal project details. The goal is to stay clear for search engines, AI systems, and human visitors while preserving privacy and security boundaries.

Engineering Timeline

2019-2021

Java Backend & Business Systems

Built the foundation around Spring Boot, MySQL, Redis, API development, and business system delivery.

2021-2023

Microservices, Middleware & Data Pipelines

Went deeper into Spring Cloud, message queues, search, caching, tracing, and complex system decomposition.

2023-2025

Cloud Native, Platformization & Automation

Focused on Kubernetes, DevOps, platform engineering, automation tools, and engineering efficiency.

2025-2026

AI Engineering, Indie Building & Long-term Works

Combining LLMs, Agent / MCP, real-time voice, indie tools, and backend engineering into systems that can run, be maintained, and accumulate over time.

Focus Areas

Backend & Platform Development

Microservices, API Gateway, Service Governance, Data Pipelines, Caching, Message Queues, System Stability, Performance Optimization, DevOps, Cloud Native

AI Engineering

LLM application engineering, agents, Function Calling / MCP, model serving, real-time voice pipelines, evaluation, and observability

Indie Tools & Products

Browser-side tools, AI workflows, rule sets, code generators, personal automation, and reusable small product experiments

Writing & Long-term Archive

Original engineering practice, tutorial series, project reviews, AI research, personal growth, and long-term public work systems

Why I Write This Blog

This blog is not about churning out content or copying what others have written.

It's more like a long-term experimental field to record real problems and solutions in technical practice, my understanding of backend and AI application delivery, experiences with hardware and automation, as well as thoughts on running, growth, and long-term thinking.

What I aim to achieve with this content:

What I Write Here

If you're interested in these topics, there's probably something useful here.

What I Believe

Technology for me is not just a work tool, but a way to understand the world, transform problems, and amplify personal capabilities.

About Me

Outside of work, I also continuously record life and thoughts. I love running, technology, and experimenting with new tools. I also like turning scattered experiences into written records. Compared to "hype", I care more about long-term accumulation; compared to "showing off", I care more about real growth.

This site itself is part of my long-term accumulation.

Contact

✉️ Email wuzikang1998@gmail.com 🐱 GitHub wdkang123 📚 CSDN w776341482 ⛏️ Juejin Juejin Profile 💼 LinkedIn wuzikang
This site records technology, and also growth.
I hope that years from now, what's left here is not just articles, but also the trail of someone continuously moving forward.
📝 Read Blog Tech Articles & Project Reviews 🔧 Explore Projects What I'm Building 🏠 Home Quick overview of this site