Buli Xing 邢步犁

Principal / Senior Product Manager

Designing trustworthy systems at scale.

Reliable for the people who depend on them. Resilient against the people who try to break them.

I design trust into AI, robotics, and marketplace systems — before the first incident, not after.

Read selected work See what I'm building

Work across:

Amazon Robotics Amazon Merchant Risk Microsoft via FiveBy Johns Hopkins SAIS

Currently shipping

I build, not just steer.

The same discipline earlier in the cycle, while the trust problem is still being defined.

North Star

In production

A market-intelligence decision layer that turns scattered signals into a focused, evidence-backed category diagnosis.

See it →

Flowreps

TestFlight beta

An on-device speech coach — pace, pauses, and fillers per rep. Audio never leaves the phone.

See it →

Selected work

Two case studies, one underlying discipline.

The patternStart with who the system has to hold up against — then design trust in structurally, not bolted on after the first incident.

Case study I · Amazon Robotics

The three questions that made Sparrow trustworthy.

01

Economic

Could the business afford to trust it?

Build cost cut, payback cleared
02

Architectural

Could the platform scale it?

First customer of a shared stack
03

Operational

Would the people run it?

Handoff to FC associates
Open the full case study →

Case study II · Amazon Merchant Risk

Build controls around actors, not accounts.

Controls-firstbefore
Rules matched against transactions.
New sellers and coordinated fraud as one class.
Enforcement blunt — punishing to the innocent.
Actor-firstafter
Start with the adversary, derive the control.
Entity-linkage separates amateur from adversary.
Sharper on abuse, lighter on legitimate sellers.
Open the full case study →