Workshop 3 · Worked examples
Each use case is a small system you could build, with the prompt that drives it and the watch-fors that come with letting an AI do part of your work.
Custom bot
Studying for a CompTIA exam over 12 weeks is hard mostly because nobody is checking in on you. A small custom bot that runs once a week — Sunday evenings — and uses your own notes to suggest next week's focus is the cheapest accountability partner you can build. Build it once, return weekly.
OpenAutomation
Help desk technicians and SOC analysts spend a real chunk of their week reading user-forwarded suspicious emails. A classifier doesn't replace the human read, but it can rank the queue: highly-likely-phishing first, marketing-noise last. Built on a Make scenario or Zapier Zap that watches a 'phishing-reports' inbox label, this is the second-most-common automation a Tier 1 builds for themselves.
OpenCustom bot
Cold outreach to people whose roles or companies you want to learn from is one of the highest-leverage moves in early career networking. It is also one of the easiest places to do badly: the templates are obvious, the AI tone is obvious, and one bad message burns the connection. A custom bot that drafts a personalized opener — and stops there — is the right shape. The bot writes; you read, edit heavily, and decide whether to send.
OpenAutomation
Tier 1 and Tier 2 help desk technicians often have a daily stand-up where they report what they worked on yesterday and what's on deck today. Writing those notes after closing 30 tickets is annoying. An automation that pulls your closed-ticket list from a sandbox or your own notes log and writes a tight stand-up paragraph saves you 10 minutes a day and makes you look more organized than you are.
OpenCustom bot
Studying flashcards by yourself has one weakness: you grade your own answers, which means 'close enough' becomes 'correct.' A custom bot that takes your spoken or typed answer to a flashcard, compares it to the canonical answer, and grades you honestly closes that loop. This is the second-most-used Workshop 3 build for participants taking CompTIA exams.
OpenCost and limits
Every free tier in this workshop ends. Claude free has message limits; ChatGPT free has rate limits; Zapier free caps at 100 tasks/month; Make free at 1,000 ops; Perplexity free has daily query caps; the OpenAI API charges per token. If your agents are useful, you will hit a wall. The decision then is which paid plan earns the cost — not all of them do, and the wrong upgrade burns $20-100 a month for a year before you notice. This is the conversation to have with AI before you click 'Subscribe.'
Open Facilitated by Lemieux Consulting. Hosted by the Urban League of Louisiana.