Bingran You
尤炳然
- Agentic Builder
- Ion Trapper
I build reliable AI systems and trapped-ion quantum experiments. The work runs on two clocks: long lab cycles measured in vacuum-pump hours, and product cycles measured in days between commits.
- 2022 — Present
University of California, Berkeley
PhD Candidate in Applied Science & Technology · Berkeley, California
Haeffner Lab
- 2018 — 2022
University of Chinese Academy of Sciences
BS in Physics, Minor in Computer Science · Beijing, China
GPA 3.95 / 4.00 · Rank 1 / 54
- SkillsBenchbenchflow-ai/skillsbench
A benchmark for evaluating how well AI agents use skills.
track:agents·01 / 05open - first-treeagent-team-foundation/first-tree
A Git-native context layer for decisions, ownership, and shared team knowledge.
track:agents·02 / 05open - DoWhizKnoWhiz/DoWhiz
Agent-native product for getting work done across email, chat, documents, and related tools.
track:agents·03 / 05open - DeepTutordeeptutor.knowhiz.us/
An AI research assistant built on Zotero for cited answers, figure and formula understanding, and multi-paper comparison.
track:agents·04 / 05open - mewsbingran-you/mews
Local GitHub notification daemon that triages your inbox and dispatches Codex or Claude Code work for allow-listed repos while you sleep.
track:agents·05 / 05open
Git-native context layer for AI agents. Pre-1.0; iterating with early teams.
Co-author on a Nature paper this year. Now: scaling and integration.
Codex + GPT-5.4 last 7 days. Most of it is agent evaluation runs.