Xiao L.

Most teams hire me when manual workflows start breaking under scale — too many tools, too much copy-paste, data that won't stay in sync. I build full-stack SaaS platforms and AI automation that make those problems go away. My recent work spans the full range. For a B2B procurement CRM, I shipped a unified inbox that merges Gmail, Slack, and WhatsApp into one threaded view, plus a Human-in-the-Loop AI quoting engine that turns messy supplier negotiations into structured, versioned PDFs. I also built a real-time IoT operations platform for an EV battery-swap network — a Next.js / React command center over Java Spring Boot microservices, tracking stations, batteries, and swap orders live at high concurrency. On the AI and automation side, I build agent pipelines and n8n workflows that parse documents, orchestrate OpenAI calls, and push results into Google Workspace, Slack, and Twilio. One example: an AI agent that reads a tender PDF, scrapes Google Maps to discover matching vendors, enriches their contact data, and launches a targeted outreach campaign — turning days of manual sourcing into minutes. The foundation comes from Alibaba and Vipshop (NYSE: VIPS), where I built Java/Spring Boot systems handling 100k+ DAU at 99.9% uptime and once cut a procurement batch process from 24+ hours to under 3. I build production-grade from day one — not something you'll have to rewrite in six months. Stack: Java/Spring Boot, Python, Node.js, Next.js, React, PostgreSQL, OpenAI/Claude APIs, RAG, n8n, AWS, Docker. I translate business goals into technical roadmaps, own delivery end to end, and keep you updated daily. If you're building something that has to work reliably at scale, let's talk.