Operators - tell me about your AI use cases

March 30, 2025

by an investor from University of Pennsylvania - The Wharton School in St. Louis, MO, USA

I'm interested in hearing how you have found success deploying AI within your company. What specific use cases have you found success with and others that have failed? What drove the success vs. the failures? What were the small steps you took to start? What do you wish you'd done sooner? How much technical talent was required to deploy? Did you outsource that technical talent or did you have it in-house?

There has been a lot of discussion about AI use cases in marketing and customer service applications, I'm curious if there are other success stories (or failures) beyond these.

Thanks!

4 Likes

10 Replies

105 Views

Replies

Reply by a professional
from University of Illinois at Urbana in Indianapolis, IN, USA

I would characterize much of what I'm reading on Searchfunder as "prompt engineering". That's useful as a foundation and, frankly, that's where I started. I bought a "prompt library" and followed the advice provided along with the prompts to get an idea of what it means to ask good questions. What you learn very quickly is to start at a high level and ask increasingly more detailed questions. Most of the AI providers have improved their "memory" capabilities which just means you can now reference the prompt replies to glean further insights. Simply stated: subsequent prompts can reference information you've already been given. Some of the best use cases, in my experience, have come from using AI prompts to come at a line if inquiry from different angles. I then upload all of that information into a "knowledge base" from which I can begin to ask questions that require a more comprehensive perspective on all the available information (within a narrowed context). Often I will use multiple AI tools to produce inputs into this process. I think this approach will ultimately provide the best value to Searchfunders. Sometimes this is called "building your own GPT"; others simply call it a "project". If you go to chatGPT and go to "Explore GPTs", you'll see dozens of specialized GPTs. You can also build your own. The banner on this page reads: "Discover and create custom versions of ChatGPT that combine instructions, extra knowledge, and any combination of skills." I've use this approach to compile information and query the custom GPT for strategy direction all the way through to project plans. I'm happy to help with building custom GPTs if there's an interest. I have a corporate technical background so that's been helpful - mostly from the standpoint of not being afraid to explore and push the boundaries. Some of the things I wish I'd done sooner include: using AI for business automation, AI capabilities for Vibe Coding (using AI to write code).

Reply by a searcher
in United Kingdom

Full data segmentation, being able to uncover the entire TAM and being able to deploy a unique message to each segment and individual. Also, automation of routine tasks - from emails, pre-call lead analysis, setting up triggers for top 100 accounts (for example announcements, someone steps in a new position etc.), identifying website visitors and many more.

+8 more replies.

Join the discussion