Eric Yin
Make Gemini better
- Role
- Software Engineer Machine Learning Engineer at Google
- Location
- Palo Alto, CA, US
- LinkedIn followers
- 500 followers
About Eric Yin
Improve Gemini post training on instruction following and general text generation quality.
Experience
Software Engineer Machine Learning Engineer
May 2025 — Present · Mountain View, CA, US
Tech Lead Gemini general Text Quality at Vertex AI. Post-training research to improve Gemini on enterprise use cases.Recent research key words:* Gemini repetitive token loss pattern debug and improvement* Improving RL performance on non-verifiable domains such as text writing quality* Autorater quality refinement for ultra complex instruction following prompts* Gemini Ultra Complex System Instruction Following - Eval set generation and refresh - Loss pattern analysis - Post-training contribution and model capability improvement.* Seamless Gemini 3 Migration - Large scale migration readiness analysis and report generation - Loss pattern and triggering condition analysis - Migration prompting guide generation
Education
Stanford University
Artificial Intelligence Professional Education
KTH Royal Institute of Technology
Exchange
2016 — 2017
Georgia Institute of Technology
Master of Science (M.S.), Computational Science and Engineering
2017 — 2019
Dalian University of Technology
Bachelor of Engineering (B.Eng.), Computational Solid Mechanics
2013 — 2017
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.