I build at the intersection of strategy, product, and AI.

Fraud strategist and product builder using AI to transform complex problems into practical systems, useful products, and smarter ways of working.

Selected impact
60%fewer unauthorized transactions
$2.5Msavings from fraud strategies
40%less login friction
3rd placeRBC Agentic AI Showcase
01 / Experience

Protecting trust with data and better decisions.

Fraud strategy, authentication, analytics, and the controls built from them.

Leading fraud controls across wires, ACH, Zelle, and non-monetary activity—turning emerging risk signals into scalable strategies, monitoring systems, and operating workflows.

  • Reduced unauthorized transactions and customer friction by 60% through authentication redesign, real-time rule optimization, and stronger control enforcement.
  • Built a behavioral case-review framework combining phone intelligence, risk signals, and historical fraud trends to improve investigation efficiency by 60%.
  • Created centralized Snowflake and Tableau reporting for fraud KPIs, Reg E and UTR service levels, and executive performance visibility.
  • Authored audit-ready procedures and playbooks that standardized fraud handling, regulatory documentation, and team training.
  • Designed AI prototypes for conversational investigations, chat-with-data analytics, and self-serve regulatory and training guidance.
  • Built an AI-assisted investigation concept that placed 3rd among 180+ RBC enterprise teams.
Tools used
SnowflakeTableauSQLThreatMetrixRSAPindropOmilia

02 / Applied AI

Agents built around the work.

Workplace agents and independent automations designed to return time and sharpen judgment.

02A / Workplace AI

AI prototypes for fraud operations.

Public-safe concepts for investigations, control strategy, and analyst enablement.

Turns a client fraud-reporting call into an investigation-ready package: key facts and participants, a fraud narrative, a documented 14-question intake, and a follow-up email that asks only what the call did not answer.

Less time documenting. More time investigating.
01Call02Extract03Narrative0414-point intake05Follow-up
60 → 10–15 minreview time per case
~$85Kestimated annual savings
3rd placeRBC Agentic AI Showcase
Transcript analysisStructured extractionGap detectionHuman review

02B / Independent automations

Independent n8n workflows.

Demo-led experiments connecting models, data, and APIs. Open one to explore its architecture.

Workplace concepts are prototypes. Savings and capacity figures are estimates based on stated review volume and time assumptions. Employer-specific data and implementation details are excluded.

03 / Products

From thesis to working products.

Three AI-assisted builds across personal finance, investing, and thoughtful gifting.

Live on iOS

01 / PERSONAL FINANCE

Finesse

Your money, but it actually watches itself.

Finesse connects accounts through Plaid, learns normal behavior, detects recurring costs, and surfaces severity-tagged anomalies—so users know what changed without reviewing every transaction.

Tools used
SwiftUIPlaidSupabaseRevenueCatAPNs
Visit Finesse
Finesse anomaly alert linked to its underlying transaction
Finesse home screen showing safe-to-spend and monthly finances
Finesse recurring subscription history and annual cost
Unusual spend caught$3,360 under forecast

UPCOMING

Maya profileMaya's birthday7 days away
WHY THIS FITS MAYA

“She carries a notebook everywhere. Something tactile and personal feels right.”

Personalized journal gift suggestionCurated serving board gift suggestionHomeware gift suggestion
5 thoughtful picks ready
Latest product · live web app

02 / AI + CONSUMER

EmptyHanded

Never show up empty handed again.

An AI gifting concierge that remembers the people and occasions that matter, learns what each person loves, and sends thoughtful recommendations before the date sneaks up.

Tools used
Next.jsReactGeminiTailwind CSSVercel
Visit EmptyHanded
Live web product

03 / AI + FINANCE

Stockee

The market, translated.

Stockee closes the knowledge gap for newer investors by combining price action and technicals, fundamentals, SEC filings, earnings-call transcripts, institutional and insider activity, congressional trades, prediction markets, YouTube, and Reddit—then explaining what the signals mean.

Tools used
Next.jsSupabaseGeminiPlaidStripePostHogVercel
Visit Stockee
NVDANVIDIASNAPSHOT
$182.14+2.8% today
STOCKEE TRANSLATIONGrowth is still doing the heavy lifting.

Revenue momentum remains strong, but the valuation leaves less room for an earnings miss.

EARNINGSBeat3 of 4 quarters
INSTITUTIONSAddingNet accumulation
SENTIMENTMixedWatch volatility
RESEARCH INPUTS
Market dataTechnicalsFundamentalsSEC filingsEarnings callsInstitutionsInsider tradesCongressPrediction marketsYouTubeReddit
04 / Writing

Proof beyond the job titles.

Recognition and ideas shared while building.

01Agentic product demo

Finesse acts on financial insights—with your approval

A preview of an agentic finance experience that explains the next best action, works through iMessage, and keeps every decision under the user’s control.

Watch the Finesse agent demo
02Enterprise recognition

3rd place at RBC’s Agentic AI Showcase

Selected from 180+ submissions to represent City National Bank, then presented an AI-enabled fraud use case across RBC.

Read the showcase post
03Product thinking

Why money software should work in the background

The product thesis behind Finesse: learn normal behavior, surface what changed, and make the interface available when useful—not mandatory.

Read the Finesse post
04Build in public

Making investment research legible for beginners

How Stockee brings fragmented market signals together and translates them into plain-English context instead of stock picks.

Read the Stockee post
Core skills

Built to connect the dots.

Fraud expertise, analytical depth, product leadership, and AI-assisted prototyping.

01

Fraud & risk

Fraud strategy & decisioningAccount takeover & authenticationPayments fraud: wires, ACH & ZelleFraud rules & controlsPortfolio monitoring & case reviewBehavioral and social-engineering analysisThreatMetrix, RSA, Pindrop & OmiliaReg E, Reg Z, NACHA & UTR
02

Data & analytics

SQL: Snowflake, SQL Server & OraclePython and R for analysisTableau and advanced ExcelKPI frameworks & reportingData pipeline designExperimentation & statistical analysisDecision trees & behavioral featuresData visualization & storytelling
03

AI & automation

AI-assisted product prototypingAgent & workflow designPrompt engineeringRAG & vector searchn8n automationsLLM evaluation & iterationConversational and voice AIHuman review & document intelligence
04

Product & delivery

Product strategy & managementRequirements, use cases & prioritizationCross-functional executionAI-assisted web and mobile buildsAPI & webhook integrationsSupabase, Plaid, Twilio & StripeAgile, Scrum, Jira & ConfluenceSalesforce, UAT & regression testingProcess mapping, procedures & playbooksStakeholder management & analyst training