At Camerry Ice Cream, I visited 1,048 retail outlets across seven Kerala districts and spoke with owners, dealers, and distributors about stock, shelf visibility, and lapsed accounts.
I built a verified dataset from that fieldwork and found that half of the dealers checked carried no branding. I tested the gap by district, distributor, and visit pattern, then gave management a costed plan using existing visits and flagged outdated records for correction.
For GAP, I built an app that analyzes Reddit and Hacker News discussions to assess whether a consumer category is worth entering. Its scoring weighs pain prevalence, competitor weakness, volume, and momentum, and compares brand negativity and complaint phrases to find positioning gaps.
I also built an LSTM price-forecasting app and reported its performance against a naive baseline, concluding that gains came from features rather than architecture. My toolkit includes Python, SQL, Excel, and market research methods.
