✨ TLDR
Helped a global retail analytics company transform their legacy dashboards into a GenUI AI agent. Users went from navigating complex dashboards to just asking questions and getting visual answers.
MVP shipped in 1 week. Presented at NRF Singapore 2025 and NRF US 2026. Now piloting with some of the largest retail brands in the world.
🎯 The Problem
The client is a retail analytics company that’s been around for 7-8 years. They serve major retail brands across India and the US — we’re talking billion-dollar companies, top 5 global retailers.
Here’s the thing: they had great data. Trend predictions, market insights, competitive intelligence. The kind of stuff their customers desperately need.
But nobody was using it.
Why? Their dashboards were too complex. Users had to click through multiple screens, understand the data structure, and figure out how to get what they needed. Most people just… gave up.
đź’ˇ The Solution
Build an AI agent that responds with interactive visual UI — not just text.
The idea: instead of navigating dashboards, just ask what you want.
- “What are the top trending colors in the US market?”
- “What’s Brand X doing in Europe right now?”
- “Show me emerging micro-aesthetics in women’s knitwear for FW26”
And the agent responds with charts, images, trend visuals — the actual stuff designers and buyers need to see, not walls of text.
This was early 2025, before Generative UI was a common pattern.
🛠️ What I Built
Week 1: MVP
Got a working prototype up in about a week.
After MVP: Evals
This is where the real work happened. The team was shipping features, but users weren’t happy with the responses. Classic AI product problem.
I worked with their subject matter experts (designers, retail consultants) to:
- Create a golden dataset of expected Q&A pairs
- Build evaluation pipelines to score agent responses
- Set up a system where they could systematically improve the agent
Evals are underrated. You can’t improve what you don’t measure.
Ongoing: Advisory
After the initial engagement, I stayed on in an advisory role — jumping in when their team hit blockers or needed architecture guidance.
đź“Š Results
- Presented at NRF Singapore 2025 and NRF US 2026 (the biggest retail tech conferences)
- Currently piloting with top global retailers
- Team can now iterate on the agent systematically using evals
🧠What I Learned
Evals > prompts. Everyone wants to tweak prompts. But building evaluation infrastructure pays off 10x more. Once you can measure quality, improvement becomes systematic instead of random.
Domain expertise matters. I’m not a fashion expert. Working with their SMEs to understand what “good” looks like was essential. The golden dataset we built together was more valuable than any prompt I could write.
Ship fast, then stay. The MVP got us to a demo. But the real value came from the months of iteration after — helping the team build the muscle to improve the product themselves.
đź”— Related
- AI Consulting — More about my consulting work