|
It's remarkable to think about the amount of work that went into my tasks back in 2017 when I worked as an analyst for a retail company. Much of my analysis was designed to assist store managers and the eCommerce team, ensuring the success of the company's omnichannel strategy. Typically, my day started with location data obtained from market research firms. I'd input this data into our database, use SQL for querying, run the relevant R or Python code, and then send the report to the respective requester. In summary, this entire process took about 2-3 days and required writing at least 50 lines of code. Now, with the advancements in LLMs, I believe traditional methods of analyzing location data will soon be obsolete, especially when it comes to deriving straightforward insights. So, I've enhanced GPT-4 (In-context learning) with the location data from over 75,000 US locations, allowing me to query it directly. (Background) I launched Zipscore last year, aiming to provide small businesses with easy access to location data. Many large firms charge as much as $2000 for such a service, not including consultation fees. While Zipscore didn't initially gain the traction I'd hoped for, I've persisted in developing it. Recently, I've felt that it's the right time to pivot and transform it into an AI Co-Pilot for location data and market analysis. I'd greatly appreciate any feedback you might have. Thank you! |