Jonathan Gu
I build systems that fuse economic theory with production ML at Instacart — marketplace optimization, availability forecasting, and growth experimentation that handle billions in GMV. PhD Economics from UCLA, BA Statistics & Economics from Berkeley.
Causal Inference
A/B design, guardrails, heterogeneous treatment effects, and ROI modeling for product decisions.
ML Systems
Reproducible training pipelines, feature stores, eval harnesses, and observability for production models.
Economic Mechanisms
Auction design, multi-agent incentives, and policy robustness that optimize long-run marketplace health.
Product & Strategy
Hypothesis framing, experiment design, and clear decision memos that turn analysis into shipped impact.
Instacart
San Francisco- Built item availability ML system serving real-time predictions across millions of SKUs with pinpoint accuracy
- Designed causal measurement frameworks that capture long-term effects without sacrificing decision speed
- Shipped growth experiments and forecasting systems handling billions in annual GMV
"First Month Free" — From Signups to Retention
Full-funnel subscription experiment with day-of-week modeling, decision ranges, and automated guardrails.
Reproducible Training for Availability Forecasting
Standardized model training via JSON configs tied to feature tables, with eval harnesses and on-call alerting.
Auctions & Mechanism Design
Search auction mechanisms and bidding strategies with emphasis on long-run value and incentive alignment.
PhD, Economics
UCLA — 2014–2020
BA, Statistics & Economics
UC Berkeley — 2007–2011
我們能做的，就是將自己的事做到最好，其它，交給命運吧。
All we can do is do our own work to the best of our ability. The rest, leave to fate.
— 顧鳴高