Professional Summary

I’m Jackie Jiaqi Yin, an ML Scientist at OpenAI focused on B2B revenue forecasting and strategic modeling. My day-to-day work centers on understanding business demand, improving forecast quality, and turning modeling results into decisions that can support revenue growth and better planning.

You may also find me online as Jackie Yin, Jiaqi Yin, or 殷佳祺.

My background also spans the broader AI stack. Before OpenAI, I led AI product work in Microsoft Copilot Notebooks, where I worked on context engineering, evaluation, tool usage, and grounded AI experiences. Across my career, I have combined machine learning, experimentation, forecasting, and product thinking to build systems that are useful in practice, not just accurate on paper.

Alongside my core role, I stay close to self-driven AI work: open-source tooling, research-reading systems, data mining utilities, and collaborations in areas such as health AI and aging. That mix of business modeling, applied AI, and independent technical curiosity is what defines my work.

Current Work

OpenAI

January 2026 – Present

I work on B2B revenue forecasting and strategic modeling, with an emphasis on understanding demand signals, identifying growth drivers, and translating noisy business data into decisions that matter.

What I Bring

  • Machine learning and forecasting for business-critical planning problems
  • Strategic modeling that connects data signals to growth decisions
  • A product mindset shaped by prior AI system and evaluation work
  • A strong interest in building practical AI tools outside the core job

Microsoft — Senior Applied Scientist

Aug 2020 – Jan 2026 | Redmond, WA

Copilot Notebooks (2025 – 2026)
Led AI product and evaluation work focused on data curation, context engineering, prompt reasoning, tool usage, and LLM-as-judge evaluation for scenarios without ground truth. As an early senior member, I helped shape the AI foundations behind grounded, reliable notebook experiences.

Amplify AI & Business Insights (2020 – 2025)
Led recommendation systems for Microsoft 365 support, including a two-tower model and Thompson sampling bandit approach. I also served as the core scientist behind a forecasting service deployed in Dynamics 365 for internal and external customers, and helped define the ML infrastructure and experimentation frameworks that supported those systems.

Key Themes Across Roles:
Applied AI system design, context engineering, LLM evaluation, recommender systems, forecasting, experimentation, and business-focused modeling.

Areas of Expertise

  • Revenue Forecasting & Strategic Modeling – B2B demand understanding, growth planning, and decision support
  • AI Product Development – context engineering, prompt design, tool use, and grounded experiences
  • AI Evaluation – scalable measurement for systems without clean ground truth
  • Retrieval-Augmented Generation – grounded knowledge access and synthesis
  • Experimentation & Metrics – product iteration, A/B testing, and business insight
  • Recommendation Systems – personalization and adaptive decision algorithms

Education

  • M.S. in Computer Science, Georgia Institute of Technology, GPA 4.0/4.0 (May 2025)
  • Ph.D. in Biostatistics, University of Washington, (Jul 2020)
  • B.S. in Mathematics & Applied Mathematics, Zhiyuan Honors Class, Shanghai Jiaotong University, GPA Rank: 2nd (Jul 2015)

Beyond Work

Outside of work, I like building small open-source projects that help people read papers, work with data, or automate repetitive AI tasks. I also enjoy exploring new ideas across research, product design, and the kinds of technical problems that sit just outside the boundary of my day job.