Hej!

I’m Jingwen (Lambo) Zhuang, a software engineer based in Stockholm. I did my B.Sc. in Software Engineering at Zhejiang University of Technology in Hangzhou, China, came to Sweden as an exchange student, and stayed for a master’s in Interactive Media Technology at KTH. I’m currently at Ericsson, where I wrote my master thesis on how network conditions affect the experience of talking to AI agents in real time, and I’m now building a voice + vision AI assistant on AR glasses as a research intern. What I enjoy most is building real-time systems that people actually interact with, and then measuring whether they actually work. Outside of work I’m a tech nerd. I play many games (Minecraft, Assassin’s Creed, Cyberpunk 2077), travel whenever I can, do analog photography sometimes, listen to a lot of 80’s music, and I’m quite fond of Disney’s productions.

📍 Stockholm, Sweden 🇸🇪

💻 Skills

AI / Agents: LLM APIs (Gemini Live, OpenAI), real-time voice/multimodal agent pipelines (Pipecat, LiveKit), agent evaluation & controlled user studies
Backend: Python, Node.js/Express, GraphQL, REST, MongoDB, WebSocket/WebRTC, Docker, Linux, CI/CD (GitHub Actions)
Frontend & Mobile: TypeScript, React/Next.js, Vue, Swift/SwiftUI
Game & XR: Unity, Godot, C#, C++, OpenGL
Other: Figma, AI-assisted development (Claude Code, Copilot)

🎓 Educational Journey

KTH Royal Institute of Technology · MSc, Interactive Media Technology · Stockholm, Sweden · 2024–2026 (graduating Sep 2026)
Blekinge Institute of Technology · Exchange year, Software Engineering · Karlskrona, Sweden · 2023–2024
Zhejiang University of Technology · B.Sc., Software Engineering · Hangzhou, China · 2020–2024

🛠️ Experience

Research Intern at Ericsson, Consumer & IndustryLab, Stockholm (Jun 2026 – present)
- Developing a real-time multimodal AI assistant on AR glasses (RayNeo X3 Pro, Android): live voice and video are streamed to a Gemini Live–based agent that guides users through hands-on tasks.
- Designing a controlled user study to quantify how network conditions affect agent responsiveness and user experience, reusing a self-built network emulation platform for reproducible conditions.
Master Thesis at Ericsson, Consumer & IndustryLab, Stockholm (Jan 2026 – Jun 2026)
- Developed a VR voice-agent system for studying conversational AI quality of experience (QoE): built a Unity client and adapted a Pipecat-based agent backend (VAD→STT→LLM→TTS) for real-time voice interaction in VR.
- Developed a Linux-based experiment orchestration platform on top of netem, enabling reproducible user studies under controlled network conditions (latency, bandwidth, packet loss).
- Conducted a controlled user study (N = 19) on how network degradation affects conversational agent QoE; found that uplink packet loss hurt the experience disproportionately by corrupting speech-to-text input to the LLM.
Teaching Assistant at KTH, DH2642 Interaction Programming and the Dynamic Web (Nov 2025 – Jan 2026)
- Supported students on a 100+ student course building React/Vue applications (state management with Redux and MobX); ran quizzes and coached project teams through the full development cycle.
President of ZJUT iOS Club (2022-2024)
- Founded and led the club, providing a platform for students at ZJUT interested in iOS development.
Member of Jinghong Network Student Association of ZJUT (2020-2021)
- Took part in various web projects with Vue.js and Go/Gin.

📃 Awards & Languages

Winner, Apple WWDC Swift Student Challenge (2021, San Francisco, USA)
3rd Prize, 7th CCCC Mobile Application Innovation Contest Finale (2022, Hangzhou, China)
English (fluent, IELTS 7.5) · Chinese (native) · Swedish (basic, actively learning)

📄 CV

Download my CV (PDF)