Skip to content

Technical

What is Magic to you in Software Development?

Arthur C. Clarke famously said: "Any sufficiently advanced technology is indistinguishable from magic." What is your "this is magic" in software development? Put another way what is something you currently don't understand but would like to?

As an older developer I'm struggling to fully understand LLMs and agents. How do you train one? Why are some agents/LLM models so much better than others? I'm curious to hear what others struggle with.

Clarke's Three Laws: https://en.wikipedia.org/wiki/Clarke%27s_three_laws

Everyone and anyone are welcome to join as long as you are kind, supportive, and respectful of others. Zoom link will be posted at 12pm MDT.

P.S. - The featured image is from Claude Opus 4.8. In my personal experence Claude is the best for development but horrible at images. The output from Claude about the image made me laugh and reminds me of when a kid draws a picture and you have to sneakly ask them what it really is.

Here's an illustration for your post—a wizard's crystal ball displaying a neural network and glowing nodes, resting on an open spellbook whose pages show streams of code. The Clarke quote sits at the top, with your discussion prompt anchoring the bottom, and magical sparks scattered throughout to tie the "technology as magic" theme together.

A bad Claude generated image that is supposed by be crystal ball over a spell book

Context Engineering: Prompting Is Not Enough

This week, let's talk about context engineering: the practical work around AI systems that goes beyond writing a single prompt. Good results often come from better context shaping, retrieval choices, evaluation loops, and clear guardrails, not just "better wording."

Discussion starters:

  • What has had the biggest impact for you: prompt structure, system instructions, retrieval quality, tool use, or evals?
  • Where do AI-assisted workflows still break down in real teams (code review, debugging, docs, testing, handoffs)?
  • What's one habit or pattern you've adopted that made AI outputs more reliable?

Everyone and anyone is welcome to join as long as you are kind, supportive, and respectful of others. Zoom link will be posted at 12pm MDT.

In honour of Amii's Upper Bound conference happening this week, the May 19th, 2026 chat will be about AI. Often we limit how much AI chatter is allowed but not today! Anything AI related. Bring your AI hopes, worries, success stories, failures, etc.

Everyone and anyone are welcome to join as long as you are kind, supportive, and respectful of others. Zoom link will be posted at 12pm MDT.

P.S. - The Weekly Dev Chat will be at the Upper Bound Mega Meetup (May 19th, 5pm to 8pm, downtown library). Swing by and say hi.  A Upper Bound ticket is NOT required to attend the Mega Meetup.

https://www.meetup.com/edmontonunlimited/events/314599992

https://www.upperbound.ai/