AAOS — Data Science Engineer
Track AI research signals
Collect updates from selected AI labs, AI for Science startups, research feeds, and technical blogs.
AI Research Intelligence + Scientific Agents
ScientificLoop tracks AI agents, AI for Science, scientific workflow automation, and model infrastructure, then turns the strongest signals into cited notes and small public-safe prototypes.
Map signals to scientific workflows
Separate general model news from signals relevant to agents, RAG, evaluation, simulations, robotic labs, and scientific decision support.
Turn strong signals into prototypes
Use small demos and notebooks to test whether a pattern can help scientific navigation, formulation, evidence review, or workflow automation.
Write cited working notes
Convert useful findings into public-safe posts with source links, open questions, and a ScientificLoop angle.
Daily workflow
AI research radar → post → prototype
The publishing workflow starts with automated source collection, then adds human judgment: why the signal matters, how it affects scientific agents, and whether it suggests a ScientificLoop demo.
npm run daily:radar
npm run new:post -- --title "Scientific agent workflow note"Agent and AI systems
4 lu-zhang-site
lu-zhang-site
This site — academic homepage with structured data layer
Astro static site driven by curated JSON data extracted from a master research record. 99 publications (97 SCI + 2 book chapters), 28 patents (25 issued + 3 applications), 51 conference presentations (38 abstracts + 13 invited talks).
Live
↗ Live site
meditation-study
meditation-study
Personal meditation learning platform
Web app for studying a meditation teacher-training course. Built around session JSON, polished script generation, and TTS audio playback. Daily-use product.
Live
↗ Live site
guangnian-energy-agent
guangnian-energy-agent
光年新能源智能体 — Azure-deployed RAG assistant for new-energy content
Deployed bilingual RAG assistant built on Azure OpenAI and Azure AI Search patterns, with a Python aiohttp backend and lightweight web UI. Used as a practical AI x energy demo and workflow prototype.
Live demo
↗ Live site ⌗ Source
rfb-formulation-agent
rfb-formulation-agent
AI scientist for redox flow battery formulation
A pre-v0.1 vertical-domain AI agent prototype designed to support redox-flow-battery electrolyte formulation. It aims to translate 14+ years of flow-battery chemistry intuition into structured reasoning, retrieval, and validation workflows.
In progress · pre-v0.1
More tools and experiments live on github.com/rockyzl — primarily personal Claude Code skills, RAG / agent prototypes, and research scripts. They are not productized and not listed here.
Industrial roles
Mercury Insurance — Senior Data Scientist
Cincinnati Insurance — Data Scientist