5+ years of frontend engineering, most recently focused on AI/Agent product frontends. At my current company I'm one of the earliest engineers on a digital-twin + AI life-sciences platform, owning the frontend from scratch: React 19 + Vite + TailwindCSS, SSE-based streaming chat, agent "thinking process" visualization, tool-call rendering, and markdown rendering — took it from 0 through PoC/demo to an internal beta in about 3 months.
Before that, 4.5 years at a large telecom company building B2B enterprise platforms (video terminal management, cloud-PC/VDI monitoring, network device management). Highlights:
- Two full framework migrations: Angular 9 → Vue2, then Vue2 → Vue3, including iframe-based micro-frontend dual-stack coexistence while de-risking heavily coupled legacy code
- Sole frontend owner on a cloud-PC operations/monitoring platform: built the entire Vue2 foundation (routing, permissions, i18n, component library) from zero, plus a JSON-config-driven reporting system covering 15–20+ reports with zero-code extensibility
- Data visualization: Echarts monitoring dashboards, AntV G6 network topology, responsive big-screen layouts across varied hardware
- Shipped a company-wide CI/code-quality dashboard (SonarQube + Jenkins) used by 100+ teams, earned the top performance rating and a promotion for it
- Backfilled a Grafana plugin (custom Echarts flow visualization) after self-teaching React under deadline pressure
Comfortable owning frontend architecture end-to-end: from tooling/framework selection to shipping. I've been practicing AI-assisted development since ChatGPT's early days, using Cursor / Claude Code daily inside a disciplined spec-driven workflow (SDD + OpenSpec), treating AI as leverage for velocity, not a substitute for engineering judgment.
Open to full-time remote roles; willing to relocate for the right team.
zs1m · · focus · HN ↗
Remote: Yes, actively seeking overseas remote roles
Willing to relocate: Yes, for the right opportunity
Technologies: React 19, TypeScript, Vite, TailwindCSS, Zustand, TanStack Query, React Router, shadcn/ui, SSE/streaming UI, Vue 2/3, Angular (migration), Three.js/React Three Fiber, Echarts, AntV G6, Grafana plugin dev, i18n, SonarQube + Jenkins CI
Résumé/CV: <a href="https://contrails.space/en" rel="nofollow">https://contrails.space/en (blog) | <a href="https://github.com/zS1m" rel="nofollow">https://github.com/zS1m
Email: mail@contrails.space
5+ years of frontend engineering, most recently focused on AI/Agent product frontends. At my current company I'm one of the earliest engineers on a digital-twin + AI life-sciences platform, owning the frontend from scratch: React 19 + Vite + TailwindCSS, SSE-based streaming chat, agent "thinking process" visualization, tool-call rendering, and markdown rendering — took it from 0 through PoC/demo to an internal beta in about 3 months.
Before that, 4.5 years at a large telecom company building B2B enterprise platforms (video terminal management, cloud-PC/VDI monitoring, network device management). Highlights:
- Two full framework migrations: Angular 9 → Vue2, then Vue2 → Vue3, including iframe-based micro-frontend dual-stack coexistence while de-risking heavily coupled legacy code - Sole frontend owner on a cloud-PC operations/monitoring platform: built the entire Vue2 foundation (routing, permissions, i18n, component library) from zero, plus a JSON-config-driven reporting system covering 15–20+ reports with zero-code extensibility - Data visualization: Echarts monitoring dashboards, AntV G6 network topology, responsive big-screen layouts across varied hardware - Shipped a company-wide CI/code-quality dashboard (SonarQube + Jenkins) used by 100+ teams, earned the top performance rating and a promotion for it - Backfilled a Grafana plugin (custom Echarts flow visualization) after self-teaching React under deadline pressure
Comfortable owning frontend architecture end-to-end: from tooling/framework selection to shipping. I've been practicing AI-assisted development since ChatGPT's early days, using Cursor / Claude Code daily inside a disciplined spec-driven workflow (SDD + OpenSpec), treating AI as leverage for velocity, not a substitute for engineering judgment.
Open to full-time remote roles; willing to relocate for the right team.