Marketing Analytics · End to End · ServiceNow
CampaignAttribute IQ
A multi-touch attribution engine built to give every marketing touchpoint the credit it deserves
In B2B marketing, most teams credit only the last campaign before a deal closes — erasing every webinar, email, and content piece that moved the prospect forward. CampaignAttribute IQ was built to change that. Starting from nothing, we designed and delivered the company's first multi-touch attribution model, giving marketing teams a truthful picture of which campaigns were actually driving pipeline.
Multi-touch attribution
Markov chain modeling
Data modeling
Data science collaboration
Cross-functional leadership
Go-to-market
Read full case study →
Product Design & Analytics · ServiceNow
EffectTrack
A measurement framework built to show marketing where its budget is actually working
Every marketing team runs programs — but most can't say which ones moved a deal forward. EffectTrack was designed from the ground up to answer that question, with a five-tab framework built around the specific decisions each user group needed to make.
Product design
Wireframing
User research
Data visualization
Stakeholder management
Tableau
Read full case study →
Generative AI · Agentic Systems · UCONN
AgentLens
A GenAI agent built to surface pricing insights from raw marketplace data
Marketplace analysts spend hours manually processing pricing data and building client decks. AgentLens automates that end-to-end — raw Excel in, formatted insight presentation out — handling the first 70-80% of an analyst's work automatically.
GenAI agent designDIFY platformPrompt engineeringAgentic workflows
Read full case study →
Decentralized AI · Energy Forecasting · UCONN
Subnet 183
Building a decentralized AI energy forecasting platform from the ground up
Most AI today is built and controlled by a handful of large organizations — and hard real-world problems like energy demand forecasting rarely get open, participatory solutions. Subnet 183 is a live, open forecasting platform we built from scratch — where student miners compete to predict real ISO-NE grid electricity demand, scored and ranked automatically. As PM Lead I designed the architecture, aligned the team, and presented at two university AI events. Then competed as a miner — ranking #1 out of 76 participants over a 7-day window.
PM leadershipDecentralized AIML modelingArchitecture designBittensor
Read full case study →
