Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm Yiqun Li (Richard Li), a passionate AI/ML engineer and senior full-stack architect with 8 years of experience delivering AI-native platforms and high-concurrency systems. I specialize in AI/LLM application development, Retrieval-Augmented Generation (RAG), and big data processing, and I have a proven track record of leading end-to-end product lifecycles from ideation to production. I navigate smoothly between research and production, combining hands-on data engineering, model deployment, and frontend-backend integration. My toolkit includes Python, PyTorch, Redis, Kubernetes, and cloud-native infra, plus front-end frameworks like React/Vue. I thrive in fast-moving, high-traffic environments and am driven by building scalable, reliable systems that empower teams and users alike.…I'm Yiqun Li (Richard Li), a passionate AI/ML engineer and senior full-stack architect with 8 years of experience delivering AI-native platforms and high-concurrency systems. I specialize in AI/LLM application development, Retrieval-Augmented Generation (RAG), and big data processing, and I have a proven track record of leading end-to-end product lifecycles from ideation to production. I navigate smoothly between research and production, combining hands-on data engineering, model deployment, and frontend-backend integration. My toolkit includes Python, PyTorch, Redis, Kubernetes, and cloud-native infra, plus front-end frameworks like React/Vue. I thrive in fast-moving, high-traffic environments and am driven by building scalable, reliable systems that empower teams and users alike.

Yiqun Li (Richard Li)

AI Engineer, Full Stack Developer, Back-End Developer, +3





I'm Yiqun Li (Richard Li), a passionate AI/ML engineer and senior full-stack architect with 8 years of experience delivering AI-native platforms and high-concurrency systems. I specialize in AI/LLM application development, Retrieval-Augmented Generation (RAG), and big data processing, and I have a proven track record of leading end-to-end product lifecycles from ideation to production. I navigate smoothly between research and production, combining hands-on data engineering, model deployment, and frontend-backend integration. My toolkit includes Python, PyTorch, Redis, Kubernetes, and cloud-native infra, plus front-end frameworks like React/Vue. I thrive in fast-moving, high-traffic environments and am driven by building scalable, reliable systems that empower teams and users alike.…I'm Yiqun Li (Richard Li), a passionate AI/ML engineer and senior full-stack architect with 8 years of experience delivering AI-native platforms and high-concurrency systems. I specialize in AI/LLM application development, Retrieval-Augmented Generation (RAG), and big data processing, and I have a proven track record of leading end-to-end product lifecycles from ideation to production. I navigate smoothly between research and production, combining hands-on data engineering, model deployment, and frontend-backend integration. My toolkit includes Python, PyTorch, Redis, Kubernetes, and cloud-native infra, plus front-end frameworks like React/Vue. I thrive in fast-moving, high-traffic environments and am driven by building scalable, reliable systems that empower teams and users alike.

Available to hire

I’m Yiqun Li (Richard Li), a passionate AI/ML engineer and senior full-stack architect with 8 years of experience delivering AI-native platforms and high-concurrency systems. I specialize in AI/LLM application development, Retrieval-Augmented Generation (RAG), and big data processing, and I have a proven track record of leading end-to-end product lifecycles from ideation to production.

I navigate smoothly between research and production, combining hands-on data engineering, model deployment, and frontend-backend integration. My toolkit includes Python, PyTorch, Redis, Kubernetes, and cloud-native infra, plus front-end frameworks like React/Vue. I thrive in fast-moving, high-traffic environments and am driven by building scalable, reliable systems that empower teams and users alike.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

English

Fluent

Work Experience

Founding Engineer (AI & Full Stack) at JuziLab

March 1, 2024 - Present

Led the entire lifecycle of an AI-native vertical research platform—from RAG tuning and PDF image extraction to full-stack delivery. Implemented a two-stage retrieval pipeline (BM25 plus dense vector embeddings) with a cross-encoder re-ranker to boost precision, and added semantic chunking to improve retrieval quality. Enabled concurrent querying across 50+ papers in a NotebookLM-style workflow for literature analysis. Built an in-app PDF image extraction tool using a custom YOLOv5 model, with dataset of 1,000–2,000 images and training details; designing an AI-assisted editor with a real-time Markdown-to-PDF engine and WebSocket streaming for a cursor-like writing experience.

Tech Lead / Core Architect at Shuzhan Tech (ChatX)

January 1, 2023 - February 1, 2024

Led architecture and core development for a multi-modal AIGC content-creation platform, solving high-cost and concurrency challenges early in AI applications. Designed 'Cloud LLM + Local SD' dual-engine system to reduce API costs, integrating GPT-3.5 for text and a private Stable Diffusion inference cluster with LoRA and Prompt Template Middleware. Achieved ~90% reduction in per-image generation cost and full autonomy at scale for core logic. Implemented a distributed asynchronous Redis-based task queue with a 1:1 Worker-to-GPU binding, enabling traffic shaving and 0 task loss/VRAM OOM incidents under high concurrency. Improved UX with WebSocket streaming and a UGC distribution flow for one-click sharing.

Backend Engineer at Collection KA (NFT Marketplace)

December 1, 2020 - October 1, 2022

Architected a high-concurrency NFT trading platform capable of 200k+ QPS during flash sales, delivering zero downtime and sub-50ms core API latency. Led Huawei Cloud (Kubernetes)-based cloud-native infra with auto-scaling, reducing routine infrastructure costs by ~30%. Implemented a trading engine using Distributed Locks (Redisson) and eventual consistency to resolve concurrency in blind box openings, synthesis, and C2C trades, enabling millions in GMV with zero asset loss and data integrity.

Data Engineer / Information System Engineer at IfreeComm

June 1, 2018 - May 1, 2020

Designed and built an enterprise data platform on Hadoop/Spark; automated ETL with Python and Airflow, reducing data processing time and enabling same-day data availability. Improved query performance by introducing ClickHouse for OLAP and optimizing SQL to cut report loading from 30+ seconds to under 3 seconds. Built BI dashboards with Vue.js and ECharts for interactive data visualization and decision support.

Campus Information System Engineer at State University of New York (SUNY)

June 1, 2017 - May 1, 2018

Managed daily operations for the Enterprise Information System; optimized ticket workflows and automated dispatch to significantly improve response speed and resolution rates. Independently designed and launched an 'Equipment Asset Management' module on top of the legacy system, including backend schemas and frontend dynamic inventory views, enabling real-time tracking and digital asset management for teaching equipment.

Data Engineer / Information System Engineer at freeComm

June 1, 2018 - May 1, 2020

Designed and built an enterprise data platform on Hadoop/Spark. Implemented automated ETL pipelines in Python with Airflow, improved query performance by adopting ClickHouse for OLAP, and delivered same-day data availability. Built BI dashboards with Vue.js and ECharts for interactive data visualization and decision support.