daily revenue reached through TikTok Shop Ads systems
Greater Seattle Area | ML Systems | Commerce Ads
Sitian Lu
Staff Machine Learning Engineer / Tech Lead
I build large-scale ML systems that turn product, user, and marketplace signals into measurable business growth.
Ads ranking
Retrieval
Commerce intelligence
Engineering leadership
$3M+
daily revenue
+55%
retrieval lift
+35%
ranking lift
Impact Snapshot
Scale made legible.
TTMall ad revenue lift from retrieval improvements
TTMall ad revenue lift from ranking architecture
candidates served through low-latency ANN retrieval
team led across ML, product, ads, and commerce delivery
GMV Max campaign penetration reached through optimization
Mission
Turn ambiguity into systems that perform.
My work is about finding the right signals, designing reliable ML infrastructure, and aligning technical choices with product and business outcomes. The craft is not only model quality. It is judgment, latency, data, operational discipline, and the confidence to lead teams through unclear terrain.
Journey
A path from platform craft to marketplace intelligence.
Purdue University
Computer EngineeringComputer Engineering foundation with a 3.96/4.0 GPA, grounding the later work in systems thinking and engineering rigor.
Throughline Engineering rigor became the pattern: understand the system, then improve the signal.
Foundation
Salesforce
Web platform & analyticsBuilt reusable web components on AEM and implemented analytics frameworks across Salesforce.com experiences.
Throughline Platform craft and measurement became a habit: build reusable surfaces, then instrument the experience.
Web platform | analytics
Amazon
Retail servicesDesigned large-scale retail services and distinctive shopping experiences for customer subsegments.
Throughline Retail systems added the marketplace context that now shows up in commerce ads work.
Retail services | operations
TikTok
Shop Ads ML systemsLeads ML systems for TikTok Shop Ads delivery, including retrieval, ranking, Carousel Ads, GMV Max optimization, and engineering quality practices.
Throughline The path converged: marketplace context, ads delivery, ML systems, and technical leadership at global scale.
Staff ML | ads delivery | leadershipSystems I Build
Technical depth, translated for humans.
Signature Achievements
Proof points with architecture behind them.
Each milestone represents both a technical build and a business outcome: retrieval quality, ranking intelligence, delivery systems, team leadership, and measurable revenue.
Personal Layer
The same person, different lenses.
A good bio page should flex depending on who is reading. Use the lens control to shift emphasis between collaboration, friendship, and partnership without changing the core identity.