Executive Profile

Amelia Quan

LinkedIn
Product & Technology Executive in Enterprise AI

15+ years turning new technology into products people use and results that show up on the P&L, from founding product teams at startups to leading enterprise AI at global companies.

I also advise leadership teams on agentic AI strategy, enterprise enablement, workflow transformation and responsible AI adoption.

Discuss an AI Engagement

Now

I lead product and technology for CBRE Capital Markets: I set AI and product strategy, own the portfolio and roadmap, and build the platforms, data foundations and governance that let AI scale.

What I delivered

I grew GenAI SaaS ARR 180% at Cellebrite, grew AI revenue ~5× at Clearview AI and drove 5× booking conversion at Sabre.

How I lead

I build forward-deployed teams, set clear governance and run a delivery process that ties every AI investment to a business result.

How I got here

Four chapters, one direction.

  1. 2012–2016Digital commerceTravelocity · Hilton

    I learned conversion, experimentation and personalization at scale.

    PersonalizationA/B testingSEOMerchandisingAnalyticsPartner platforms
  2. 2016–2020Building & commercializing MLSabre · iDonate · Topgolf

    I built and sold ML personalization to a global hotel network at Sabre, then shipped ML platforms and computer vision.

    ML personalizationComputer visionML scoringPlatform modernizationAPI integrations
  3. 2020–2023Scaling AI businessesRollKall · Clearview AI

    I grew AI revenue, retention and trusted public-sector adoption.

    AI matchingFacial recognition AIAI pricingAI licensingResponsible AICompliance frameworks
  4. 2023–nowGenAI & agentic AICellebrite · Channel Fusion · CBRE

    I work at the frontier: I build LLM products, agentic systems with reusable skills and tool connectors, and the evaluation and governance to run them in production.

    LLMsHybrid RAGAgentic AIPrompt engineeringAI evaluationHuman-in-the-loop

Building with ML and AI since 2016, across 30-person startups and 20,000+ person enterprises.

My impact

Outcomes and KPIs I delivered

+180%
ARR growth on the GenAI SaaS line I scaled at Cellebrite
Enterprise commercialization
~5×
AI revenue growth I delivered at Clearview AI
Public-sector go-to-market
−45%
Hardware cost I cut at Topgolf
+10% revenue, +30% release velocity
5×
Booking conversion from the ML personalization I led at Sabre
Global hotel network
~90%
Faster lending workflows targeted at CBRE
Soft quotes in hours, not days
60%
Lower decision latency from the hybrid RAG workflows I built at Cellebrite
LLM-powered field analysis
26% → 65%
Platform adoption I drove at iDonate
OPEX down 15%
85%
Customer retention I delivered at RollKall
+20% ARR uplift
Outside work
RunningHow I think through hard problems.
WeightliftingRegular time in the gym, built on consistency.
Reading & writingAlways in a book, and always putting ideas on paper.
My dogsMy favorite coworkers.
Experience

From startup formation to global enterprise scale.

How I led, what I built and what changed, role by role.

CBRE
2026 – Present
Director, Product & Technology — Capital Markets

I own product and technology for Capital Markets across Investment Properties, Debt & Structured Finance and Loan Servicing.

Leadership

  • Lead forward-deployed and traditional product teams across Investment Properties, Debt & Structured Finance and Loan Servicing.
  • Built the executive AI portfolio: field AI ideas shaped into reusable components, funded leadership bets and a transparent prioritization model.
  • Shape AI strategy and operating model across the business: governance, AI cost economics, adoption and the path from proof of concept to governed workflow.

Technology

  • Designed and built an agentic producer assistant on live Snowflake and Salesforce data, built from reusable skills and connectors under one app.
  • Diagnosed enterprise property-data architecture and its identity crosswalk; set the data-readiness, lineage and identity-resolution plan for AI.
  • Led debt and servicing automation concepts: soft quotes in hours instead of days, and ~90% less effort on loan narratives and spreading QC.
  • Designed human-in-the-loop approval for every downstream action.
StackSnowflakeSalesforceAWSLLMsRAGPrompt tuningGitHub
Results I deliveredSoft quotes in hours, not days~90% faster loan narratives~90% less spreading QC
Channel Fusion
2025 – 2026
Director, Head of AI & Product Innovation — Contract

I headed AI and product innovation for AI-driven marketing services.

Leadership

  • Led AI product and engineering teams in a forward-deployed model.
  • Created an AI Acceleration Kit and a 0-to-1 product development lifecycle for discovery, prioritization and delivery.
  • Improved on-time delivery ~30%.

Technology

  • Turned customer and business needs into AI-driven marketing capabilities.
  • Delivered automated ad generation and audience segmentation.
  • Launched AI compliance validation for campaign content, cutting manual review 50%+.
StackLLMsMicrosoft AzureMicrosoft FoundryRAGPrompt tuningGitHub
Results I delivered25–40% faster campaign builds50%+ less manual review~30% better on-time delivery
Cellebrite
2023 – 2025
Director, Product Management — AI SaaS

I owned a GenAI SaaS product line for investigative analytics.

Leadership

  • Grew GenAI SaaS ARR 180% with a market-aligned roadmap.
  • Introduced packaging and tiered pricing for enterprise commercialization.
  • Owned AI quality metrics: accuracy, hallucination rate, cost efficiency, usability and cloud scalability.

Technology

  • Built LLM-powered OSINT and field-analysis workflows.
  • Shipped hybrid RAG with prompt tuning as cloud-scaled SaaS.
  • Cut decision latency 60%.
StackAWSAWS BedrockHybrid RAGPrompt tuningGitHub
Results I delivered+180% ARR growth60% lower decision latency
Clearview AI
2022 – 2023
VP, Head of Product & SLED Sales — AI Facial Recognition

I ran product and public-sector sales for an AI facial-recognition platform.

Leadership

  • Ran product alongside state, local and federal sales.
  • Grew ARR ~5× through product strategy, bundled licensing and public-sector GTM.
  • Partnered across product, engineering, sales and customers.

Technology

  • Set product strategy for proprietary facial-recognition AI.
  • Built compliance and adoption frameworks into the product.
  • Ran product feedback loops and onboarding and ethics enablement for public-safety users.
StackProprietary facial recognition models
Results I delivered~5× AI revenue growth
RollKall Technologies
2020 – 2022
VP, Head of Product & AI Innovation

I led the post-acquisition transformation of an AI workforce-matching platform.

Leadership

  • Led the post-acquisition product transformation.
  • Established product and advisory practices to support scale.
  • Raised retention to 85% and delivered 20% ARR uplift.

Technology

  • Rebuilt AI workforce matching and matching intelligence.
  • Shipped an AI-enhanced UX.
StackWeb and mobile appML matching models
Results I delivered85% retention+20% ARR
Topgolf Entertainment Group
2019 – 2020
Director, Head of Product — Interactive Experiences

I headed product for Swing Suite interactive experiences.

Leadership

  • Owned P&L and roadmap for Swing Suite.
  • Delivered 10% revenue growth.
  • Increased release velocity 30%.

Technology

  • Deployed computer-vision and ML gameplay scoring.
  • Integrated hardware and software in a physical-digital experience.
  • Cut hardware costs 45%.
StackFull Swing simulatorsComputer visionML scoringConnected hardware
Results I delivered+10% revenue−45% hardware cost+30% release velocity
iDonate
2018 – 2019
VP, Head of Product — Contract

I was the founding product hire for a donation platform.

Leadership

  • Recruited as the founding product hire; built the function, operating discipline and roadmap from the ground up.
  • Introduced Agile delivery.
  • Reduced OPEX 15%.

Technology

  • Launched an ML-enhanced donation platform.
  • Led the platform modernization.
  • Grew adoption from 26% to 65%.
StackML modelsSalesforceBlackbaudMicrosoft Dynamics
Results I delivered26% → 65% adoption−15% OPEX
Sabre Hospitality
2016 – 2018
Sr. Principal Product Manager — AI Hospitality Stack

I led the AI Hospitality Stack for a global network of hotel clients.

Leadership

  • Built the product function for the AI Hospitality Stack from the ground up.
  • Established product discipline, roadmap and delivery practices.
  • Drove strategic partner integrations.

Technology

  • Led AI personalization across a global hotel network.
  • Led the platform migration and ML-driven SynXis redesign.
  • Achieved 5× booking conversion.
StackSynXis CRSSynXis Booking EngineTripCaseML personalizationPartner APIs
Results I delivered5× booking conversion
Hilton Worldwide
2014 – 2016
Regional Manager, Product eCommerce

I led regional product eCommerce across a portfolio of Hilton hotels.

Leadership

  • Established the regional product function from the ground up: processes, roadmap and team structure.
  • Built and managed the regional team.
  • Hilton Innovation Challenge finalist.

Technology

  • Led personalization across the regional hotel portfolio.
  • Ran SEO and digital product experiments.
  • Increased revenue 16%.
StackPersonalizationSEOA/B testing
Results I delivered+16% revenue
Travelocity (Expedia)
2012 – 2014
Senior Manager, eCommerce & Product Partnerships

I led eCommerce and product partnerships.

Leadership

  • Established the product function within eCommerce & Partnerships from the ground up.
  • Built the roadmap discipline and processes behind partner enablement.

Technology

  • Built merchandising, analytics and self-service partner capabilities.
  • Used behavioral insights to improve partner enablement and digital conversion.
StackMerchandising toolsAnalyticsPartner self-service
Additional experience: DexYP (SuperMedia) — Senior Manager, Search Engine Marketing (2009 – 2010) · Bank of America — Financial Advisor (2008 – 2009) · Nationstar Mortgage — Senior Loan Officer (2007 – 2008)
Technical & Executive Scope
Agentic AIGenAI / LLM ProductsRAGPrompt EngineeringSnowflakeSalesforceAWSAzureData / Identity ArchitectureAPIs / IntegrationsAI GovernancePricing & MonetizationP&L / CapEx / OpExProduct Operating Models
Education & Credentials
M.S., Information TechnologyPenn State University
B.B.A., FinanceUniversity of North Texas
AI Product ManagementDuke University
Product Manager Level VIIPragmatic Institute
Generative AI for BusinessApplied enterprise AI training
Prompt EngineeringOpenAI API · LangChain · Pinecone
AI Lab & Projects

How I build AI, and where I have done it.

I start with the data: where it comes from, who owns it and how it moves. Then I build models and agents on top, with a person approving every action.

1 · Architecture

How I build: data first.

Four layers, with trust running through all of them. Gold notes show what I built at each layer. The demos and projects below map back to these layers.

Trust layer
Layer 1ExperienceWhere people use AIWeb, CRM, mobile and embedded copilots.
What I built
  • Mobile and web assistant for field teams
  • One app on top of shared skills
Layer 2OrchestrationWhat happens, in what orderRouting, tool selection, workflow state, policy checks.
What I built
  • Orchestrator that routes each request to agents
  • Reusable AI skills, built once, called anywhere
Layer 3AI capabilitiesThe modelsLLMs, RAG, extraction, classification, ranking, vision.
What I built
  • LLM products and hybrid RAG
  • ML personalization, matching and computer vision
Layer 4Enterprise contextThe data that makes answers rightWarehouse, CRM, documents, identity, MDM.
What I built
  • Diagnosed enterprise property-data architecture
  • Identity-resolution and lineage plan
  • Connectors to CRM, warehouse and documents
Trust layerIdentity and permissions, evaluation, audit, cost telemetry and human approval, applied to every layer above.
What I built
  • Governance tiers set by risk and reversibility
  • Human approval before any agent acts
  • Evaluation and AI cost economics
2 · Demos

See it working.

Three working demos on synthetic data. Each one exercises specific layers of the architecture above.

How it works: map the workflow, fix the data (identities, lineage, owners), connect governed sources, build reusable skills that cite their sources, ship with human approval.

Opportunity AgentSynthetic demo

Agentic reasoning path

Tap an agent for its inputs, guardrail and metric.

Data before modelsIf we cannot trace the data, we do not build on it.
Lineage on every answerEach output links to its source record and system.
Reusable skillsBuild once, govern once, call anywhere.
People approve actionsNothing is written or sent without a human.
3 · Projects I led

Products and transformations I owned.

What I built, the data behind it and the result, grouped by type. Layer tags map each project to the architecture. Open any card for detail.

Agentic AI & GenAI

Agents and LLM products in production.
CRE vertical
Agentic Producer Intelligence

I built agents on live Snowflake and Salesforce data. Every recommendation cites its source record.

OrchestrationEnterprise context
Agentic AI
  • Reusable AI skills and connectors under one application layer.
  • Runs on web and mobile from one shared skill layer.
  • Next-best actions grounded in source data, with human approval.
Digital forensics vertical
GenAI SaaS & Hybrid RAG Decision Workflows

I shipped GenAI for investigators, grounded in case data with hybrid RAG and measured on accuracy and hallucination rate.

AI capabilitiesTrust
GenAI SaaS
  • I grew ARR 180%.
  • Hybrid RAG + prompt-tuning patterns for OSINT and field analysis.
  • I cut decision latency 60%.
MarTech vertical
AI Acceleration Kit & Forward-Deployed Model

I built a repeatable 0→1 system from use-case discovery to delivery.

AI capabilitiesExperience
Applied AI
  • 0→1 product development lifecycle for discovery, prioritization and delivery.
  • Forward-deployed teams: embedded discovery wired straight into delivery.
  • On-time delivery up ~30%.

Data & operating model

The foundations that make AI trustworthy and fundable.
CRE vertical
Enterprise Data & Identity Foundation

I assessed enterprise property data and its identity crosswalk, and defined which data AI can trust.

Enterprise contextTrust
Data / MDM
  • I diagnosed enterprise property-data architecture and its identity crosswalk, and defined the readiness, lineage and identity-resolution plan for AI.
  • Set lineage, ownership and identity-resolution standards.
  • Kept data and workflows independent of any model vendor.
CRE vertical
AI Portfolio & Workflow Transformation

I designed the operating model for AI intake, reuse and cost, with data readiness as an intake criterion.

OrchestrationTrust
Transformation
  • Field ideas → capability themes → reusable components.
  • Funded leadership bets with a transparent prioritization model.
  • Governance tiers, CapEx/OpEx visibility, human-in-the-loop design.

Applied ML

Models inside customer-facing products.
Hospitality tech vertical
ML Personalization at Global Hospitality Scale

I led ML personalization across a global hotel network.

AI capabilitiesExperience
ML / Enterprise
  • I drove 5× booking conversion in targeted experiences.
  • SynXis redesign and partner integrations.
Sports & entertainment vertical
Swing Suite Computer Vision / ML Scoring

I shipped computer-vision scoring in a live consumer product.

AI capabilitiesExperience
Computer Vision
  • I delivered 10% revenue growth.
  • I cut hardware cost 45%.
  • I increased release velocity 30%.
Nonprofit fintech vertical
ML-Enhanced Donation Platform

I launched ML-enabled giving on a modernized platform.

AI capabilities
ML / Platform
  • I grew adoption from 26% to 65%.
  • I reduced OPEX 15%.
  • Built product function and roadmap as founding product hire.
Public safety vertical
AI-Enabled Workforce Matching

I rebuilt AI matching and UX after an acquisition.

AI capabilitiesExperience
ML / Marketplace
  • I raised retention to 85%.
  • 20% ARR uplift.
  • Rebuilt product/advisory practices to support scale.

AI commercialization

Turning AI into revenue and P&L impact across three industries.
Public sector vertical
AI Commercialization & Public-Sector Adoption

I led AI product and public-sector GTM, with compliance built in.

AI capabilitiesTrust
AI / GTM
  • I grew revenue ~5×.
  • Onboarded public-safety users across mission-critical contexts.
  • Ethics enablement built alongside growth.
MarTech vertical
AI Marketing Services for Clients

I turned customer needs into AI marketing capabilities: ad generation, segmentation and compliance checks.

AI capabilitiesExperience
AI / Revenue
  • Automated ad generation and audience segmentation.
  • AI compliance validation on campaign content.
  • Campaign builds 25–40% faster; manual review down 50%+.
CRE vertical
Debt & Servicing Automation

I led AI automation concepts for debt and servicing, with time and cost targets tied to the P&L.

OrchestrationAI capabilities
AI / P&L
  • Soft quotes in hours instead of days.
  • Loan narrative drafting ~90% faster.
  • ~90% less financial-spreading QC effort.
AI Advisory & Transformation

From AI experimentation to enterprise adoption.

I help leadership teams identify the right AI opportunities, design scalable agentic capabilities, enable their people and build the governance and operating model required to scale.

What I offer

AI Leadership Accelerator

A full engagement, not just a kit. I assess where your organization stands on AI, train your leaders and teams, and set up the enablement to scale it.

Includes my accelerator kit of templates, scorecards and playbooks.

  • EvaluateAI readiness, data and lineage, and your use-case portfolio.
  • TrainLeaders and teams, from 50 people to 1,000+.
  • EnableOperating model, governance and adoption that lasts.
Discuss an AI Transformation EngagementExecutive advisory, workshops, enterprise enablement and transformation engagements available.
Ways I work with companies

Engagements built around the outcome you need.

Start with leadership alignment, a focused opportunity sprint, enterprise enablement or ongoing advisory. Each engagement is tailored to the organization, its data and the workflows that matter most.

Executive AI & Agentic Workshop

Leadership alignment, agentic AI education, workflow opportunities and a prioritized opportunity map.

AI Leadership Accelerator

Organizational readiness, role-based training, enablement, governance and adoption designed to scale.

Agentic AI Opportunity Sprint

Identify high-value workflows, assess data readiness, determine agent vs. copilot vs. traditional software, and build a prioritized implementation roadmap.

Fractional AI Transformation Advisory

Ongoing executive support for AI strategy, portfolio decisions, operating model, governance, vendors and execution.

AI Enablement & Training Led

People have to work differently.

01

Product & engineering operating discipline

An AI-ready product lifecycle I built and ran.

  • Discovery starts with data readiness and lineage
  • Evaluation gates for accuracy, hallucination and cost
  • Human-in-the-loop release criteria
  • Forward-deployed teams shipping from GitHub
02

Enterprise & startup AI enablement

50 → 1,000+ users trained

  • From 50-person startup teams to 1,000+ user enterprise rollouts
  • Proprietary AI tools, Claude and ChatGPT
  • Focused on daily use, not pilots
03

AI onboarding & stewardship

Onboarding gets people started right. Stewardship keeps AI safe and useful after launch.

  • Usage and data policies
  • A named owner for each AI capability
  • Ethics enablement
  • Feedback loops on adoption and quality
AI Transformation Model

My AI transformation method.

Six steps I run in order, from scattered use-case demand to governed capability.

01

Discover

Start from the job, not the tool.

Map the workflow, the friction and today's baseline.

  • Workflow and decision-point mapping
  • Baseline time, quality, cost and adoption
02

Structure

Turn scattered asks into reusable patterns.

Group use cases into shared capabilities and data needs.

  • Identity and source-of-truth mapping
  • Reuse before custom build
03

Compose

Build the smallest architecture that works.

Use rules, retrieval, models and agents only where each adds value.

  • RAG on governed enterprise data
  • Human approval at consequential actions
04

Evaluate

Measure the product, not just the model.

Test AI quality and the new workflow against the baseline.

  • Groundedness, accuracy, hallucination
  • Cost per task and business outcome
05

Govern

Set controls by risk and reversibility.

Permissions, audit and approval based on what the system can do.

  • Informational, recommend and act tiers
  • Audit, telemetry and escalation
06

Scale

Productize what repeats.

Scale reusable capabilities and retire what does not pay off.

  • Shared skill registry and platform
  • Training, adoption and portfolio economics