Experiments – AI Pioneers Path

Write complex spreadsheet formulas in Google Docs

Google Sheets: a ‘smart’ multi-tab document I have a spreadsheet with one tab for each quarter (e.g. “2025Q1”, “2025Q2”), and each tab contains detailed financial data about sales, revenue, costs, profits for that quarter. This handles multi-currency billing and accounting, and recognition of sales in one quarter when closed in a different quarter etc (all…
Read more

Write a presentation/conference talk by speaking

Two challenges in conference talks The reality of writing a conference talk is that it takes a professional about 8x as long to prepare the talk as it takes to deliver it – i.e. a 1 hour talk requires approximately 1 full day of work to create. Day-to-day presentations (e.g. all-hands meetings, e.g. new sales…
Read more

AI-driven Customer Discovery to find P/MF

For startups that haven’t broken-through into scaleup territory yet, the single most important task is to find Product/Market Fit (P/MF). Eric Ries’s Lean Startup gives a great starter, and Rob Fitzpatrick’s The Mom Test gives a fast crash-course in the essentials. The hardest next step for first-time founders is often: how do I find the…
Read more

Personalised, high-quality, Xmas cards for $1.50 each

I wanted something different for Christmas cards this year – not just ‘nice’ cards, but for each recipient to receive something truly unique to them, that speaks to our personal shared history. They had to be physically printed, real cards – not cop-out ‘e-cards’ (charged at outrageous 99.9% profit margins). Process Details / prompts For…
Read more

AI-CODING: Using Claude to fix a half-written codebase

Anthropic (OpenAI’s rival) has been having a lot of problems recently – their servers keep crashing, their web interface keeps popping-up messages saying they are “downgrading all users” due to performance problems on their servers, features keep being remotely disabled by Anthropic, etc. The guys are having a hard time. But a deeply problematic side-effect…
Read more

ChatGPT part 1: create a 3D view of a 2D map from a videogame

Background The original Zelda computer games were based on large 2D maps, where the map design was critical for various puzzles, upgrades, secrets, and plot advancement. These were hand-crafted every tile, and you explored them one small segment at a time. As a player, you build up in your head a visual mental model of…
Read more

Review code-commits by Engineering staff

This is part of a larger project I’m working on with others to build more advanced, AI-augmented, Learning-and-Development / Professional-Development solutions for Software Engineers. Our goal is to help individuals (humans) improve their craft of writing software – going from “Junior Engineer” to “Senior Engineer” to “Staff Engineer” to “Principal Engineer” (and increasing their salary…
Read more

Writing a complaint email … professionally

Summary When something goes wrong with a product, and there’s no-one to complain to – but you have colleagues, or a mailing list, or a network of peers that might be able to help – what do you do? Send a Tweet, Post to Facebook, or Write an Email in anger and … as they…
Read more

Fact check something in a 1-hour video without watching the whole thing

Background: Approach Here’s the LinkedIn post original: https://www.linkedin.com/feed/update/urn:li:activity:7206986218545233920/ Here’s the YouTube video that OP claimed to take her claims from: https://www.youtube.com/live/Pig9WbMN1lQ Fortunately … that YouTube video has already been transcribed (I guess by Google?): My default learned (human) solution: My AI-centric (ChatGPT) solution: Prompt used: “In this conversation, is there a point where someone says…
Read more

Create new scalable vector icons/images from text descriptions

Usefulness: Practical/Productive Activities Techniques Discoveries Details / Example Here’s a “sheep” 😀 Comments: …and here’s the same sheep after manually tweaking it: Here’s the prompt and the output: Prompt …to get it to also fix the lack of G, I had to add: ChatGPT-4 response …and after me correcting the LLM’s weak use of SVG…
Read more

Category: Experiments